Upload
buidien
View
241
Download
0
Embed Size (px)
Citation preview
University of Groningen
Thermally reversible thermoset materials based on the chemical modification of alternatingaliphatic polyketonesAraya Hermosilla, Rodrigo Andrés
IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite fromit. Please check the document version below.
Document VersionPublisher's PDF, also known as Version of record
Publication date:2016
Link to publication in University of Groningen/UMCG research database
Citation for published version (APA):Araya Hermosilla, R. A. (2016). Thermally reversible thermoset materials based on the chemicalmodification of alternating aliphatic polyketones [Groningen]: University of Groningen
CopyrightOther than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of theauthor(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).
Take-down policyIf you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediatelyand investigate your claim.
Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons thenumber of authors shown on this cover page is limited to 10 maximum.
Download date: 03-11-2018
i
Thermally reversible thermoset materials
based on the chemical modification of
alternating aliphatic polyketones
PhD thesis
to obtain the degree of PhD at the University of Groningen on the authority of the
Rector Magnificus Prof. E. Sterken and in accordance with
the decision by the College of Deans.
This thesis will be defended in public on
Monday 17 October 2016 at 16.15 hours
By
Rodrigo Andrés Araya Hermosilla
born on 11 May 1977 in Santiago, Chile
ii
Supervisor Prof. F. Picchioni Assessment committee Prof. K. U. Loos Prof. H. J. Heeres Prof. E. Soto
iii
Dedicado a mi amada familia
Dedicated to my beloved family
iv
Contents
1 Introduction .................................................................................................... 1
1.1 Thermosetting polymers .......................................................................................................... 1
1.2 Reversible thermoset polymers ................................................................................................ 4
1.3 Self-healing polymer systems .................................................................................................. 5
1.4 Aim of the thesis .................................................................................................................... 11
1.5 References .............................................................................................................................. 13
2 Reversible polymer networks containing covalent and hydrogen
bonding interactions ....................................................................................15
2.1 Introduction ............................................................................................................................ 16
2.2 Experimental section .............................................................................................................. 19
2.3 Results and discussion ........................................................................................................... 21
2.4 Conclusions ............................................................................................................................ 30
2.5 References .............................................................................................................................. 30
3 Intrinsic self-healing thermosets through covalent and hydrogen
bonding interactions ....................................................................................33
3.1 Introduction ............................................................................................................................ 34
3.2 Experimental section .............................................................................................................. 36
3.3 Results and discussion ........................................................................................................... 39
3.4 Conclusions ............................................................................................................................ 51
3.5 References .............................................................................................................................. 52
4 Thermally reversible rubber-toughened thermoset networks via Diels-
Alder chemistry ............................................................................................55
4.1 Introduction ............................................................................................................................ 56
4.2 Experimental section .............................................................................................................. 58
4.3 Results and discussion ........................................................................................................... 61
4.4 Conclusions ............................................................................................................................ 73
4.5 References .............................................................................................................................. 73
v
5 Exfoliation and stabilization of MWCNTs in a thermoplastic pyrrole-
containing matrix assisted by hydrogen bonds .........................................76
5.1 Introduction ............................................................................................................................ 77
5.2 Experimental section .............................................................................................................. 79
5.3 Results and discussion ........................................................................................................... 83
5.4 Conclusions ............................................................................................................................ 92
5.5 References .............................................................................................................................. 92
6 Electrically-responsive thermoset nanocomposite based on Diels-Alder
chemistry .......................................................................................................95
6.1 Introduction ............................................................................................................................ 96
6.2 Experimental section .............................................................................................................. 98
6.3 Results and discussion ......................................................................................................... 102
6.4 Conclusions .......................................................................................................................... 112
6.5 References ............................................................................................................................ 113
7 Appendix .....................................................................................................117
7.1 Summary .............................................................................................................................. 117
7.2 Samenvatting ........................................................................................................................ 119
7.3 List of publications .............................................................................................................. 122
7.4 Contribution to international conferences ............................................................................ 123
7.5 Acknowledgements .............................................................................................................. 124
vi
1
1 Introduction
1.1 Thermosetting polymers
Thermosetting polymers constitute an important class of materials used for coatings,
adhesives, and electrical insulators, among several other applications [1]. Thermosets, unlike
thermoplastics, are characterized by a curing reaction that transforms two or more (liquid or
paste [2]) components into a solid network structure. Due to this network, thermosets offer
many superior properties compared to thermoplastics such as mechanical strength,
dimensional stability at elevated temperature and solvent resistance. Overall, those features
can be simply tuned by adjusting the crosslink density of the thermoset network, which in
turn results in a wide range of applications [3]. In this chapter, the history of polyketones,
their chemical modification routes and potential use as a thermosets will be briefly discussed
and framed in the general context of thermosetting polymers.
Since the development of synthetic polymers like Bakelite (phenol-formaldehydes in 1909)
many efforts have been spent at improving plastic features (e.g. strength, toughness,
recyclability, etc). During the early twenties, all efforts were essentially based on a “trial and
error” approach without any real definition of the “macromolecule” concept. It was at the end
of the twenties that this concept appeared with the work of Hermann Staudinger (1920), who
suggested the existence of macromolecules. After the twenties, polymers like PE
(polyethylene in 1940) started to replace natural materials due to several advantages of
synthetic polymers in several conventional applications (e.g. light weight, low cost, easiness
of processing). These advantages quickly widespread the use of synthetic polymers in
technological applications likes fabrics, coatings, structural materials, etc. It is during this
period (1941) that Farbenfabriken Bayer synthesized polyketone (PK) by using free-radical
polymerization (FRP) of ethylene and carbon monoxide [4]. In this first attempt, PK was
synthesized under extreme conditions (400 bar and 250 oC) and characterized as a random
distributed polymer (non-alternating). During the next 10 years, more research efforts were
spent in order to optimize the polymerization process. Indeed, Badische Anilin und Soda-
Fabrik Aktiengesellschaft (1951) synthesized a perfect alternating PK using K2Ni(CN)4 as
catalyst (Figure 1.1) [5]. Thirty years later, researcher from Shell improved the
polymerization process of PK using a bidentate-Pd2+
phosphine catalyst. Remarkably, the use
2
of this catalyst enlarged the polymerization rate of PK (6 kg PK/g Pd-1
hr -1
) under mild
conditions of temperature and pressure (50-60 bar and 70-135 o C) [6].
Figure 1.1 Alternating ethylene carbon monoxide polyketone [7].
During the nineties, several research articles have been published dealing, among others, with
a wide spreading range of potential application for PK. This stemmed from the industrial
scale by which PK can be produced and mainly consisted in the mixing with organic
additives and curing at high temperature [8, 9]. Nowadays, the chemical modification of PK
with amine compounds (the so-called Paal-Knorr reaction) has been gaining a lot of attention.
This is related to the tolerance of this reaction towards many functional groups, particularly
sterically hindered aminoes [10] (Figure 1.2).
Figure 1.2 Schematic representation of the chemical modification of polyketone via Paal-Knorr
reaction with amine compounds.
This synthetic pathway offers several advantages like high yield under relatively mild
condition (100 oC) and reaction kinetics (4 h) even without any catalyst, with water as the
only by-product. Notably, the easy chemical modification and low production costs allowed
the preparation of polymeric surfactants and nano-emulsions used for wood adhesives
application [11]. In 2009, Zhang and his co-workers published the first attempt for the
H3CCH3
O
O
CO +
n
C C
H
H H
H
K2Ni(CN)4
3
chemical modification of PK with furan amino-subtituted compounds aiming at preparing
thermoset polymer networks able to undergo reversible reactions via Diels-Alder (DA)
chemistry. The furan groups grafted directly on the PK backbone chain allowed the formation
of a three-dimensional network structure after being cross-linked with aromatic bis-
maleimide. The material indeed formed a thermally reversible and self-healing thermoset
polymer by means of DA and retro-DA sequence [12]. The properties of the resulting product
could be easily tuned by changing the crosslink density.
The cross-linking density in thermoset systems allows tailoring many aspects on their
thermo-mechanical properties. However, as the cross-linking density is increased, some
positive and negative effects can appear. For instance, highly cross-linked systems typically
display good mechanical properties such as hardness and modulus, and also remarkable
solvent resistance. Contrarily, highly cross-linked densities also give poor elongation, low
impact strength and toughness. Usually, a low cross-linked polymer system behaves as a
flexible material whereas a highly cross-linked one behaves as a brittle system. A trade-off
between high and low cross-linking density values is then necessary in order to achieve the
required properties for any given application. An example of DMTA thermograms for PK-
furan containing thermosets with different crosslink densities is displayed in Figure 1.3.
Figure 1.3 Loss (Δ) and storage (□) modulus behavior in DMTA cycles for PK50 functionalized with
furan groups (A) at different crosslink densities: (a) 1:1, (b) 1:0.5 and (c) 1:0.25 molar ratios between
furan and bis-maleimide. The softening point, taken as tan (δ) of the respective samples, is displayed
in (B) [13].
BA
4
It is clear that at higher furan/bis-maleimide molar ratios (i.e. at lower cross-linking density
values), the softening point is also diminished, thus indicating a less stiff material.
1.2 Reversible thermoset polymers
Polymer materials are nowadays essential in our society. In almost any single application,
polymers have found spots as structural or multifunctional materials. As the demand is
growing, the depleting of natural sources (e.g. petroleum) makes urgent the creation of
effective solutions on these concerns. Overcoming this huge disequilibrium between the
production of synthetic polymers and the depletion of natural resources, recyclable materials
have emerged as suitable solutions. Typically, polymers degrade with the use and age due to
mechanical stress. For instance, in an early stage of damage, unnoticed micro-cracks appear
which can lead to macro levels of damage and the failure of the whole system. The re-join of
cracks in conventional polymer systems can be achieved in different ways, like patching with
the same material or adhesives. However, the original mechanical properties of the system
will never be completely recovered. This is particularly true for thermosets, where the
chemical bonds between the chains can factually not be broken (which would allow re-
processing of the material) without any detrimental effect on the C-C bonds along the
backbone [14]. On the contrary, reversible polymeric networks can be continuously repaired
just using external stimulus like heat, pressure and light. In any case, the reversible process of
polymerization, branching or cross-linking allows in principle the recovery of the original
mechanical properties. Such reversible properties usually depend on the cleavage and
reconnection of covalent bonds. However non-covalent supramolecular interactions like π-π
staking and hydrogen bonding are also able to perform the reconnection of a severed join
under the influence of an external stimulus [15].
Reversible thermoset polymers represent a breakthrough in the general field of renewable
materials. For instance, upon thermal stimulus, thermally reversible thermosets are able to
disconnect joint functional groups and reconnect them after cooling. This is a great advantage
since polymers or even monomers can be recovered after the end of service life of a certain
material. Besides, when a reversible system undergoes a catastrophic failure, the re-join of
functional groups can recover the damaged region, thus factually extending in time the
function and service life of the material.
5
The concept of a reversible network is the conceptual basis of reworkability as well as self-
healing polymers, which will be discussed in the next section.
1.3 Self-healing polymer systems
Repair strategies for self-healing systems are commonly relying on active repairing
approaches. As a damage event occurs, external repairing agents (extrinsic approach)
embedded in a polymer matrix or the inherent functionality of certain moieties from the
polymer (intrinsic approach) will be activated in order to begin the process of repairing. This
can take place either using the damage as a trigger or using an external stimulus for healing
(Figure 1.4). Intrinsic systems, avoiding the use of any additive, would be in principle
preferable over extrinsic ones. However, the engineering and device of an intrinsic
autonomous system is still a biological attribute difficult to homologate in synthetic systems.
Figure 1.4 Extrinsic and intrinsic self-healing chemistry of polymer networks [16].
Intrinsic self-healing systems, as mentioned above, rely on special attributes of functional
groups and the nature of the polymer matrix. In general, these kinds of systems are able to
repair cracks upon external stimulus. Autonomous intrinsic self-repairing systems are so far a
difficult task for material engineering and just few examples have been mentioned which
mostly relay on reversible hydrogen bonding interactions [17-20]. As it was addressed,
intrinsic self-healing systems are based on molecular mechanisms for the repairing process.
6
The next paragraphs will briefly define the mechanisms involved in physical, chemical and
supramolecular interactions currently used for intrinsic self-healing systems.
Physical self-healing
One of the most common procedures for reversible physical interactions is through thermal
activation of the polymer chains. This was firstly studied by Wool and O’Connor for
thermoplastic materials [21]. In that work, they pointed out that crack healing proceeds
through five different stages. During thermal activation, a crack edge will start to undergo: a)
surface rearrangement that depends on the nature of the polymer material, b) the approach of
crack edges that will depend on the molecular density of polymer chains in the damage
region, c) wetting, d) diffusion that controls the recovery of mechanical properties and e)
randomization of the last conformation and entanglement of polymer chains (Figure 1.5). In a
real case, crack healing is carried out by heating the sample above its Tg to ensure reptation of
the molecules. If the reptation of the polymer chains is not achieved, a low pressure is used to
connect crack edges. After this, the mechanical strength is recovered thanks to the inter-
diffusion and re-entanglement of polymer chains. As a result of this procedure of welding
(which relies on the same steps but using higher activation energies), the mechanical strength
and toughness of cracked edges are usually recovered. The recovery degree (i.e. the degree at
which the properties can be recovered with respect to the original ones) crucially depends on
the period of the heat treatment. Prolonged exposure to healing stimuli (such as heat) could
actually result in relatively low recovery degree due to obvious side effects. Therefore, the
functionalization of polymers with active functional groups (e.g. reversible covalent or
hydrogen bonding interactions) able to recover damage in a short period of time and with low
energy could play a crucial role in order to decrease the time of exposure of the material to
healing/damaging conditions (e.g. high temperature).
As previously discussed, self-healing mechanisms are meant to re-crosslink damaged regions
on polymer networks either by physical or chemical interactions. The latter relies on certain
functional groups pending from the polymer chains and able to perform the re-connection of
polymer networks after been severed.
7
Figure 1.5 Schematic representation of physical self-heling mechanisms [22].
Chemical self-healing
In the case of reversible covalent bonds, a reversible mechanism is involved in order to
achieve the reconnection and diffusion of polymer chains at damage interfaces. However, the
process of reconnection of covalent chemical bonds usually requires relatively high levels of
energy in order to ensure the complete recovery of damage. Contrarily, supramolecular self-
healing is based on the link of supramolecular interactions that experience high mobility at
low levels of energy that ensures the faster recovery of a damaged region. In the next two
sections, both concepts commonly classified as “reversible covalent and supramolecular
networks” will be addressed in order to clarify the chemistry involved behind both concepts.
Reversible covalent network formation
One of the most emblematic examples of reversible covalent bonding is the Diels-Alder (DA)
cycloaddition reaction [23]. The reaction take place between C-C bonds and is typically in
high yield due to a good stereochemistry. Commonly, electrochemically activated diene (4π)
and dienophile (2π) compounds react to form a cycloadduct (4+2) (Figure 1.6).
8
Figure 1.6 Schematic representation of Diels-Alder cycloaddition between cyclopentadiene and
maleic anhydride.
Two stereochemically different products are mostly formed: the so-called endo and exo DA
adducts. The endo conformation is the kinetic product, the exo conformation is the
thermodynamically favourable product thanks to the less steric hindrance. As a special
feature, this kind of reaction is reversible upon heating (cycloreversion), which is commonly
known as retro-Diels-Alder reaction (r-DA) (Figure 1.7).
Figure 1.7 Schematic representation of DA and r-DA sequence between two polymeric compounds
bearing furan and maleimide pendant groups [24].
This reaction is been widely used in self-healing polymeric systems due to several reasons:
the weaker bond between diene and dienophile compounds, compared to conventional
covalent bonds, will be primarily broken during a damage event. However, upon heating the
crack will be recovered due to segmental mobility and reconnection of the DA active groups.
Among all possible alternatives, the Diels-Alder cycloaddition between furan and maleimide
moieties is indeed one of the most exploited in thermoreversible cross-linking for several
reasons. First, DA is an equilibrium reaction influenced by temperature: covalent
furan/maleimide adducts are formed at about 50 °C (cross-linking) and broken (de-cross-
Diene
Dienophile
endo
exo
9
linking via r-DA) at about 120 °C. Furthermore, the cross-linking and de-cross-linking
process can be repeated many times with negligible degradation in the range of 50 °C – 150
°C. Finally, the strong dienic character of the furan ring and the high reactivity of the
maleimide as a dienophile, ensure fast kinetics and high yields. According to literature
reviews, most of the studied thermo-reversible networks are based on DA reaction of
polymers bearing pendant furan and/or maleimide groups [15, 25-29]. For instance, a system
based on alternating aliphatic polyketones with pending furan groups and cross-linked with
1,1’-(methylenedi-4,1-phenylene)bis-maleimide was proposed as a promising low cost and
efficient thermally reversible system (Figure 1.8). As mentioned above, alternating aliphatic
polyketones obtained by copolymerization of carbon monoxide, ethylene and propylene were
modified through the Paal-Knorr reaction in order to introduce furan groups directly attached
to the PK backbone chain. The thermoset obtained upon cross-linking with bis-maleimide
was indeed re-healed up to seven times and displayed quantitatively retained mechanical
properties [12].
Figure 1.8 Thermally reversible and self-healing polyketone thermoset [12].
Supramolecular network formation
Self-healing polymer networks based on supramolecular bonds have been extensively studied
[15]. One of the main advantages obtained from reversible supramolecular bonds is the
timescale for damage recovery. Hence, reversibility, directionality and sensitivity make
10
supramolecular interactions excellent candidates in the creation of self-healing materials.
However, the resulting network formation of supramolecular bonds is rather weaker than
reversible covalent bonds. For instance, self-healing polymers based on supramolecular
interactions like hydrogen bonding or π-π stacking have the ability to recover severed bonds
thanks to the sticker-like behaviour of these interactions that enable the connection and
reconnection of broken networks upon heating [30]. However, they usually exhibit low Tg
and relatively soft mechanical properties, which hinder their potential application as
structural materials. In particular, polymer networks based on hydrogen bonding can form
clusters of hydrogen moieties depending on the acidity of the hydrogen donors. It is worth
mentioning that this represents a great advantage considering that the reversible properties
can be tuned according to the amount and modes of interaction of the hydrogen donors with
its hydrogen acceptor. In short, the kind of interaction will determine how strong the material
is and how fast and effective the healing process will be. An excellent control of highly
directional interactions between hydrogen donors and acceptors can result in a strong
reversible cross-linked polymer network as is displayed in Figure 1.9.
Figure 1.9 Schematic representation of reversible multiple hydrogen bonding interactions between 2-
ureido-4-pyrimidone molecules [31].
In this example, a very well organized molecular architecture is achieved via quadruple
hydrogen bonds using (2-ureido-4-pyrimidones) building blocks which indeed display
multiple hydrogen bonding. The reversible thermal activation of the polymeric material was
found at 90 oC, when the viscosity substantially decreases, allowing the reconnection of the
hydrogen bonds. Moreover, the material indeed found application in the real world as a
reversible elastomer and was adopted by the industry as a commercial product known as
SupraPolix [20].
+
11
Self-healing polymers based on π-π stacking interactions represent a convenient approach in
reversible systems using a thermal trigger to obtain healing processes. In particular, these
kinds of systems take advantage by combining π-electron rich and π-electron poor functional
groups as can be observed in Figure 1.10.
Figure 1.10 Schematic representation of reversible π-π stacking interactions between copolyimide
and pyrenyl end-capped chains [30].
In this example, a self-healing material based on π-π stacking interactions was achieved by
the combination of a short polymeric chain containing pyrenyl end-groups (π-electron
deficient), which interact with a long molecular chain containing π-electron-rich imide
aromatic groups. The Tg of the network was tuned in order to achieve the healing process in a
range of temperature between 50 and 100 °C by adjusting the ratio between the components.
By increasing the temperature of the system, the disruption of the π-π stacking interactions
between functional groups is promoted. This allows the flow and re-entanglement of the
polymer chains at the damaged region. After cooling, the reformation of the π-π stacking re-
establishes the network and the mechanical strength of the material [32].
1.4 Aim of the thesis
This thesis is focused on the synthesis and characterization of different kinds of reversible
thermosets and thermoset nanocomposite materials by using alternating aliphatic polyketone
(PK) as raw material. Through the different chapters, fundamental knowledge is obtained
regarding the molecular design of polymers via chemical modification of PK with aliphatic
and aromatic amine compounds. The resulting thermally reversible systems are investigated
to outline the benefits for the synergistic cooperation between reversible covalent and
supramolecular interactions. Moreover, improvements regarding the mechanical
+
12
performance, reversibility, recyclability, self-healing and electrical conductivity of the
thermoset systems have been investigated including the addition of rubber particles and
nanofillers to generate suitable industrial materials.
Chapters 2 and 3 focus on the chemical modification of alternating aliphatic polyketone with
aliphatic and aromatic amine compounds using the Paal-Knorr reaction to obtain thermally
reversible polymers with relatively high glass transition temperatures (i.e. a Tg window from
100 to 185 °C). Both chapters highlight the tunable thermal properties of polymers containing
furan (Diels-Alder) and hydrogen bonding active groups directly attached to the backbone.
The reversible thermosetting of furan-functionalized polymers via a Diels-Alder and retro-
Diels-Alder sequence with bis-maleimide is investigated and systematically compared with
thermoset polymers containing both reversible covalent and supramolecular interactions.
Chapter 4 reports on the preparation of a reversible and toughened thermoset system based on
the covalent incorporation of furan-functionalized ethylene-propylene rubber into a thermoset
furan-functionalized polyketone. Analytical, thermo-mechanical and morphological studies
have been carried out to evaluate the impact strength, reworkability and recyclability of the
toughened thermoset material.
Chapter 5 focuses on the design of an engineered thermoplastic polymer containing pyrrole
units in the main chain and hydroxyl pendant groups, which help in achieving
nanocomposites containing well-distributed, exfoliated and undamaged MWCNTs. The
incorporation of MWCNTs made it possible to change the material from an insulator to a
conductive system displaying temperature sensor properties. Notably, the resistivity–
temperature profile is very reproducible which suggests the potential application of the
composite as a temperature sensor.
Chapter 6 describes the electrically-induced self-healing properties of a thermoset
nanocomposite designed by mixing furan-functionalized polyketone cross-linked with
aromatic bis-maleimide and MWCNTs via Diels-Alder (DA) reversible cycloaddition. The
incorporation of MWCNTs increases the material modulus and allowed electrical conduction.
It is also demonstrated that above the thermal stability of the composite, the retro-DA process
is triggered reducing the mechanical properties of the material due to the decoupling of DA
adducts. Nevertheless, the mechanical properties are completely recovered after the sample is
healed through electrical resistive heating.
13
1.5 References
1. Ratna D. Handbook of thermoset resins: ISmithers, 2009.
2. Delehanty J. M, Moss I, Westall S, Dowden T. Room temperature curable silicone
elastomer composition; EP20070824749 (EP2089465 B1).
3. Deng S, Djukic L, Paton R, Ye L. Thermoplastic-epoxy interactions and their potential
applications in joining composite structures - A review. Composites Part A-Applied Science
and Manufacturing 2015; 68:121-132.
4. Ballauf F, Bayer, O., Teichmann, L. A process for preparing hoehermolekularen addition
products of lower olefins and carbon monoxide 1941 (DE863711 C).
5. Reppe W. and Magin, A. Production of ketonic bodies. 1951 (US2577208).
6. Drent E. Process for the preparation of polyketones 1984 (EP 0121965 A2).
7. Hammoud Hassan N. Synthesis of perfectly alternating carbon monoxide/olefin
polyketones using a sulfonated diphosphine catalyst system. PhD thesis, University of
Groningen 2009.
8. Auerbach AB, Hanley SJ. Stabilized polyketone polymers 1992 (US5079340 A).
9. Van Der Heide E. Thermoset resin 1998 (US5719260 A).
10. Zhang Y, Broekhuis AA, Stuart MCA, Picchioni F. Polymeric Amines by chemical
modifications of alternating aliphatic polyketones. J Appl Polym Sci 2008; 107 (1): 262-271.
11. Zhang Y, Broekhuis AA, Picchioni F. Aqueous polymer emulsions by chemical
modifications of thermosetting alternating polyketones. J Appl Polym Sci 2007; 106
(5):3237-3247.
12. Zhang Y, Broekhuis AA, Picchioni F. Thermally Self-Healing Polymeric Materials: The
Next Step to Recycling Thermoset Polymers?. Macromolecules 2009; 42(6):1906-1912.
13. Toncelli C, De Reus DC, Picchioni F, Broekhuis AA. Properties of Reversible Diels-
Alder Furan/Maleimide Polymer Networks as Function of Crosslink Density.
Macromolecular Chemistry and Physics 2012; 213(2):157-165.
14. Zhang MQ, Rong MZ, editors. Self-Healing Polymers and Polymer Composites.
Hoboken, New Jersey.: Jhon Wiley and Sons, Inc., 2011.
15. Roy N, Bruchmann B, Lehn J. DYNAMERS: dynamic polymers as self-healing
materials. Chem Soc Rev 2015; 44(11):3786-3807.
16. Billiet S, Hillewaere XKD, Teixeira RFA, Du Prez FE. Chemistry of Crosslinking
Processes for Self-Healing Polymers. Macromolecular Rapid Communications 2013;
34(4):290-309.
17. Ying H, Zhang Y, Cheng J. Dynamic urea bond for the design of reversible and self-
healing polymers. Nature Communications 2014; 5:3218.
18. Arslan M, Kiskan B, Yagci Y. Benzoxazine-Based Thermosets with Autonomous Self-
Healing Ability. Macromolecules 2015; 48(5):1329-1334.
19. Ding F, Wu S, Wang S, Xiong Y, Li Y, Li B, Deng H, Du Y, Xiao L, Shi X. A dynamic
and self-crosslinked polysaccharide hydrogel with autonomous self-healing ability. Soft
Matter 2015; 11(20):3971-3976.
20. van Gemert GML, Peeters JW, Sontjens SHM, Janssen HM, Bosman AW. Self-Healing
Supramolecular Polymers In Action. Macromolecular Chemistry and Physics 2012;
213(2):234-242.
21. Wool RP, Oconnor KM. A Theory of Crack Healing in Polymers. J Appl Phys 1981;
52(10):5953-5963.
22. Döhler D, Michael P, Binder W. Principles of Self-Healing Polymers. In: Binder WH,
editor. Weinheim, Germany: Wiley-VCH Verlag GmbH and Co., 2013. p. 66.
23. Diels O, Alder K. Synthesen in der hydroaromatischen Reihe. Justus Liebigs Ann Chem
1928; 460(1):98-122.
14
24. Sanyal A. Diels-Alder Cycloaddition-Cycloreversion: A Powerful Combo in Materials
Design. Macromolecular Chemistry and Physics 2010; 211(13):1417-1425.
25. Gandini A. The furan/maleimide Diels-Alder reaction: A versatile click-unclick tool in
macromolecular synthesis. Progress in Polymer Science 2013; 38(1):1-29.
26. Toncelli C, De Reus D, Broekhuis AA, Picchioni F. Thermoreversibility in Polymeric
Systems. In: Amendola M, Meneghetti V, editors. New York: CRC Press, 2011. p. 199.
27. Liu Y, Chuo T. Self-healing polymers based on thermally reversible Diels-Alder
chemistry. Polymer Chemistry 2013;4(7):2194-2205.
28. Long KN. The mechanics of network polymers with thermally reversible linkages. J
Mech Phys Solids 2014;63:386-411.
29. Zhang MQ, Rong MZ. Intrinsic self-healing of covalent polymers through bond
reconnection towards strength restoration. Polymer Chemistry 2013;4(18):4878-4884.
30. Yang Y, Urban MW. Self-healing polymeric materials. Chem Soc Rev
2013;42(17):7446-7467.
31. Yan X, Wang F, Zheng B, Huang F. Stimuli-responsive supramolecular polymeric
materials. Chem Soc Rev 2012;41(18):6042-6065.
32. Burattini S, Colquhoun HM, Greenland BW, Hayes W. A novel self-healing
supramolecular polymer system. Faraday Discuss 2009;143:251-264.
15
2 Reversible polymer networks containing covalent and hydrogen bonding
interactions
Abstract. Thermally reversible polymers with relatively high glass transition temperatures
(Tg) are difficult to prepare but very interesting from an application point of view. In this
chapter we present a novel reversible thermoset with tunable Tg based on the chemical
modification of aliphatic polyketones (PK) into the corresponding derivatives containing
furan (PK-FA) and/or amine (PK-DAP) groups along the backbone. The furan moieties allow
thermal setting of the polymers by the Diels-Alder (DA) and retro-DA sequence with bis-
maleimide (b-Ma), while the amine moieties allow tuning of the hydrogen bonding density.
By this combined approach, it is possible to synthesize well-defined thermosets with a wider
Tg window with respect to PK-FA alone. Indeed, after several heating cycles, the materials
showed essentially the mechanical properties of the original materials, even after samples
were broken and thermally reshaped. The incorporation of amine moieties at 20 to 60%
conversion of the 1,4-dicarbonyl units along the PK-backbone significantly increased the Tg
window from 100 °C to 185
°C. However, after each heating cycle the Tg of the materials,
detected as tan (δ) in DMTA measurements, continued to increase probably due to imine
group formation between unreacted carbonyl and amine groups. Despite this limitation, the
studied systems still display reworkability up to three thermal cycles and constitute a proof of
principle for the proposed approach, i.e. the combination of Diels-Alder and hydrogen
bonding groups towards higher Tg materials.
Keywords: Paal-Knorr chemical modification, Diels-Alder reaction, hydrogen bonding
interactions, thermoreversible polymers.
16
2.1 Introduction
Nowadays plastics are ubiquitous in the world as they are used in almost every civil and
infrastructural application. As long as new technological developments appear, cheapest
products become popular especially for common applications (e.g. plastic bags). However,
the cheapest products display in general relatively short life spans, which bring problems with
littering, management, disposal and recycling processes [1-6]. In this respect, one of the
biggest problems that involve untreated or partially treated plastic wastes is their contribution
with hazardous derivatives in aquatic and inland environments [7-10]. Thus, the reuse and
management of plastic materials has been attracting an increasing attention especially for the
definition of new strategies to reduce environmental problems like global warming [11] and
toxicity for human beings [7,12].
Since the development of synthetic polymers like Bakelite (phenol-formaldehydes in 1909)
and PE (polyethylene in 1940), many efforts have been undertaken at improving plastic
properties, often without taking into account the corresponding environmental implications.
From the 1940s, special attention has been paid to the development of cross-linked polymers
(e.g. rubbers) [13-17] and thermoset materials [18-21]. The latter are widely used in many
applications because of their remarkable mechanical properties such as relatively high
modulus (easily tailored for specific applications), solvent resistance and fracture strength.
However, as cross-linking is an irreversible process, thermosets cannot be reverted into their
un-cross-linked state and their reuse is limited to energy recovery. The latter does not fit into
the current sustainable, “cradle-to-cradle” context and does not offer any convenient, future
economic perspective as it represents a typical example of recycling while downgrading the
corresponding base material [22].
A possible solution to the general problems outlined above resides in the synthesis of
reversible cross-linked materials. Several articles have reported the ability of polymer and
composite materials to be repaired by different ways, e.g by using low molecular weight
molecules imbibed into capsules as healing agent allowing autonomic healing after damage
events (extrinsic self-repair) [23-27] or by functionalization of raw non-reworkable polymers,
for which, upon external stimuli like heat [28-33] and light [34-36], the recovery of bonding
of the functional moieties takes place (intrinsic self-repair). Against this backdrop, the use of
covalent, inter-macromolecular Diels-Alder reversible bonds represents a popular choice
17
when making allowance for the relatively fast kinetics and mild conditions at which this
reaction takes place [22, 37, 38]. In particular the furan-maleimide chemistry has been
recognized, already from the early developments in this field [30], as a viable option for the
reasons outlined above [28, 29, 31-33]. However, most of the works reported in the open
literature [22] still rely on lengthy, costly and cumbersome synthetic steps for the preparation
of the base materials, i.e. polymeric backbones displaying the presence of Diels-Alder active
groups. In this general context, our research group has recently achieved a breakthrough with
the preparation of aliphatic polyketones (PK) modified with furan groups and cross-linked
with bis-maleimide [32, 33]. This system showed an almost quantitative recovery of thermal
and mechanical behavior coupled with the possibility of finely tuning the softening point (as
measured by DMTA) in the range between room temperature and 100 °C by simply adjusting
the chemical composition of the polymeric backbone and the amount of cross-linking agent
used. Furthermore, the modification reaction for the preparation of the basis polymer (i.e.
Paal-Knorr modification of aliphatic PK with primary amines) represents a paradigmatic
example of simplicity as it proceeds in the bulk (100 °C) with relatively fast kinetics (thus
avoiding the use of any solvent or catalyst), with quantitative yields for the amino compound
(thus avoiding the necessity of any separation step) and with the formation of water as single
by-product [39, 40]. This is very attractive in a general frame of future industrial applications
[41]. Nevertheless, up to date, the upper range of softening points has been limited to
approximately 100 °C, which represents a clear drawback and limits the application field for
these materials. A possible solution to this is represented by the simultaneous presence of
different reversible interactions, able to synergistically “reinforce” each other with respect to
the thermal stability of the corresponding polymeric chains.
In this work we report our efforts aimed at preparing a thermally reversible polymer network
in which Diels-Alder and hydrogen bonds are present. To the best of our knowledge, this
represents an absolute novelty in the open literature, at least when taking into account the
systematic nature of the study. We start with the conversion of PK into PK-furan/amine
derivatives (PK-FA and PK-FA-DAP) by the Paal-Knorr reaction (Figure 2.1). This synthetic
strategy should in principle allow achieving full control of the chemical structure of the
corresponding polymers, displaying thus a tunable amount of DA and hydrogen bond active
groups [32, 33, 39]. We intend to investigate this by elemental analysis (for determination of
the conversion), 1H-NMR and FT-IR spectroscopy. The prepared polymers are then cross-
linked via DA and r-DA sequence with bis-maleimide (Figure 2.2).
18
Figure 2.1 Schematic representation of polyketone (A), PK functionalized with furfurylamine (PK-
FA) (B) and PK-FA functionalized with 1,2 diaminopropane (PK-FA-DAP) (C) by Paal-Knorr
reaction.
Figure 2.2 Schematic representation of DA and r-DA sequence of functionalized polyketone PK-FA-
DAP cross-linked with bis-maleimide. Red circles indicate hydrogen bonding interactions.
The resulting networks (Figure 2.2) should display the simultaneous presence of both
reversible interactions while the corresponding ones with only furan groups rely solely on the
19
presence of DA inter-macromolecular bonds and, as such, represent the correct reference
systems to establish the contribution, if any, of hydrogen bonding interactions. This is
checked by Differential Scanning Calorimetry (DSC) and DMTA studies.
2.2 Experimental section
Reagents
The alternating aliphatic polyketone (PK30, MW 2687 Da) presents a total olefin content of
30% of ethylene and 70% of propylene [41, 42]. Furfurylamine was freshly distillated (FA
Aldrich, ≥ 99%), while 1,2-diaminopropane (DAP Acros), butylamine (BA, Sigma Aldrich,
99%), (1,1’-(methylenedi-4,1-phenylene)bis-maleimide (Aldrich 95%), chloroform (CHCl3
Laboratory-Scan, 99.5%), dimethyl sulfoxide (DMSO, Acros, 99.7%) were purchased and
used as received.
Polyketone functionalized with Furan and amine pendant groups
The reaction between PK30, furfurylamine and 1,2-diaminopropane were carried out at
different ratios between the 1,4-dicarbonyl groups of the polyketone and the primary amine
groups of furfurylamine and 1,2-diaminopropane. The molar ratio between the reactants was
established as percentage with a maximal carbonyl conversion of 80% as previously reported
[32, 33, 39]. The chemical modifications of PK to yield PK with furan groups (PK30-FA),
amine groups (PK30-DAP) and a combination thereof (PK30-FA-DAP) were carried out in
bulk in a sealed 250 mL round-bottomed glass reactor with a reflux condenser, a U-type
anchor impeller, and an oil bath for heating. A single reaction with butylamine (BA) was
carried out in order to prepare (see below) an additional reference system. In a typical recipe,
PK30 (20 g) was preheated to the liquid state at the employed reaction temperature (110 °C),
FA and/or DAP were added dropwise to the reactor in the first 20 min. The stirring speed was
set at a constant value of 600 rpm, and the employed reaction time was 4 h [33]. The resulting
polymers were frozen with liquid nitrogen, crushed and washed 3 times with deionized Milli-
Q water to remove unreacted FA and DAP, if any. After filtering and freeze drying, light-
brown polymers were obtained as the final products. The corresponding samples are coded as
PK30-amine1 x1-amine2 x2 with xi being the mol percentage of amine with respect to carbonyl
groups in the feed. The carbonyl conversion (Xco) can be calculated by [39,40]:
20
100%%2
%%
expexpmin30
30
NCMMM
NCMXco
watereaPK
obsobsConvPK (1)
where M represents the molecular weight of polyketone before (MPK30) and after conversion
(MPK30-conv). Mamine represents the molecular weight of furfurylamine, 1,2 diaminopropane
and Mwater the molecular weight of water. Cobs and Nobs represent the weight percentage of
carbon and nitrogen derived from the elemental analysis, while Cexp and Nexp represent the
percentage expected for carbon and nitrogen resulting from theoretical calculations. The
maximum theoretical value for carbonyl conversion (THEOXco) depending on the amine
intake (g amine) in the feed can be defined as:
amine30 MPKMoles
aminegTHEOXco
(2)
The carbonyl conversion efficiency (η) can be defined as the ratio between the actual
conversion and maximum theoretical value.
XcoTHEO
Xco (3)
A maximum carbonyl conversion of 80% is assumed on the basis of published data [39, 40].
Diels-Alder reaction
The DA reaction of PK30-FA and PK30-FA-DAP (2.5 g) with bis-maleimide were carried
out in different ratios between the furan moieties and maleimide groups (Figure 2.2) using
chloroform as solvent (≈ 10 wt % polymer based on solvent) in a 100 mL round-bottomed
flask equipped with a magnetic stirrer. The reaction mixture was heated to 50 °C for 24 h to
form the polymer adduct. After reaction, the cross-linked polymers were dried at 50 °C under
vacuum overnight to remove the solvent.
Characterization
The elemental analysis was performed with an Euro EA elemental analyzer. 1H-NMR spectra
were recorded on a Varian Mercury Plus 400 MHz apparatus using CDCl3 as solvent. FT-IR
spectra were recorded using a Perkin-Elmer Spectrum 2000 equipped with a heating stage
and temperature controller. The measurements were performed in ATR modality and
“Spectrum Beer’s Law” was used as software. DSC analysis was performed on a TA-
21
Instrument DSC 2920 under N2 atmosphere. The samples for DSC were weighed (10-17 mg)
in an aluminium pan, which was then sealed. The sample was first heated from 0 to 180 °C
and then cooling to 0 °C. Four cycles were performed, and the heating and cooling rates were
set to 10 °C/min throughout the DSC measurements. The samples for DMTA analysis were
prepared by compression moulding of 450 mg of cross-linked PK30-FA / bis-maleimide and
PK30-FA-DAP / bis-maleimide into rectangular bars of 6 mm in width, 1 mm in thickness,
and 54 mm in length at 150 °C for 30 min under a pressure of 4 MPa, followed by the
thermal healing treatment at 50 °C for 24 h in an oven. The DMTA analyses of the bars were
conducted on a rheometrics scientific solid analyzer (RSA II) under an air environment using
the dual cantilever mode at an oscillation frequency of 1 Hz and a heating rate of 3 °C/min.
2.3 Results and discussion
The reactions between PK30 and furfuryl amine and/or 1,2 diaminopropane were carried out
according to different molar ratios (Table 2.1).
Table 2.1 Conversion of PK30 in presence of different ratios of furfurylamine (FA) and 1,2
diaminopropane (DAP).
Run XCO (%) a
η(%) b
PK30-BA80 78 98
PK30-DAP20 20 100
PK30-DAP40 40 100
PK30-DAP60 60 100
PK30-DAP80 79 97
PK30-FA20 19 99
PK30-FA40 39 99
PK30-FA60 58 98
PK30-FA80 76 96
PK30-FA60-DAP20 78 98
PK30-FA40-DAP40 77 96
PK30-FA20-DAP60 79 99
a % conversion of carbonyl groups.
bη conversion efficiency.
All samples with combined groups, i.e. the PK30-FA-DAP series, display a total carbonyl
conversion of about 80%, i.e. a quantitative one. Indeed, in terms of conversion efficiency, all
used amines display values around 100%, thus testifying the easiness of the reaction as well
22
as its quantitative yield for the amino compound. Slightly lower η values are obtained for FA,
in agreement with the sensitivity of this reaction to steric hindrance [39, 43].
The solubility of all obtained polymers in common organic solvents (e.g. CHCl3, acetone and
DMSO) as well as previous literature on the Paal-Knorr reaction and its sensitivity to steric
hindrance [39] indicate that only the less hindered amino group of 1,2-DAP actually reacts
with the PK backbone and thus no cross-linked product is formed. Furthermore, 13
C-NMR
spectra (not shown for brevity) lack any indication of additional imines (Schiff bases), which
should be formed upon cross-linking at such high level of carbonyl conversion. The 1H-NMR
spectra of the polymer with different pendants groups are displayed in Figure 2.3.
Figure 2.3A (reaction of PK30 with FA) shows the formation of pyrrole groups at 5.8 ppm,
the presence of CH2 units attached between the pyrrole and furan groups at 4.9 ppm and the
protons of furan moieties at 7.3, 6.2 and 5.9 ppm [32,33]. In a similar way, Figure 2.3B
(reaction of PK30 with DAP) displays the formation of pyrrole groups at 5.8 ppm, the
presence of CH2 attached to the pyrrole groups at 3.6 ppm, the CH in β position at 3.2 ppm
[39]. Finally, Figure 2.3C is showing the conversion of PK30 with both pendant groups. A
quantitative analysis of the spectra is hindered by the broad character of many peaks as well
as by their overlap for the PK30-FA-DAP. Nevertheless, at least on a qualitative level, this
agrees with the formation of the desired structure. Furthermore, spectral data obtained by FT-
IR analysis corroborate the 1H-NMR results with the appearance of pyrrole rings at 3114 and
1597 cm-1
and the furan moieties at 3150 and 737 cm-1
(Figure 2.4) [44].
As expected, when going form PK30-FA80 to PK30-FA20-DAP60 through all intermediate
composition values (i.e. from A to D in Figure 2.4) the typical furan adsorption peaks (vide
supra) decrease in intensity while the one assigned to the pyrrole rings remain unchanged.
This is in agreement with the elemental analysis data showing a total (constant) conversion of
80% for the PK30-FA-DAP series and thus a constant amount of pyrrole groups along the
polymer backbone.
All prepared polymers (before cross-linking) were characterized by DSC analysis (Figure
2.5). Polymers with only one component (either FA or DAP) display a monotonous increase
of the Tg with the amine conversion (Figure 2.5A). This is in agreement with the increased
rigidity of the backbone with increasing amount of pyrrole groups.
23
Figure 2.3 1H-NMR spectra of (A) PK30 and furfurylamine before (I) and after (II) reaction; (B)
PK30 and 1,2 diaminopropane before (I) and after (II) reaction; (C) PK30 alone (I), PK30-furan
functionalized (II) and PK30-furan after reaction with 1,2 diaminopropane (III).
Moreover, PK functionalized with amino groups (PK-DAP) systematically displays (at equal
conversion values) significantly higher Tg values than the one modified with furan ones (PK-
FA). This testifies the relevant contribution (differences are up to 90°C) of hydrogen bonding
interactions in determining the thermal properties, thus confirming the general idea of this
study (vide supra). Moreover, if one takes into account also the Tg value for PK30-BA80
8 7 6 5 4 3 2 1 0
PPM
8 7 6 5 4 3 2 1 0
8 7 6 5 4 3 2 1 0
24
(14°C), i.e. the polymer modified with butylamine (displaying the same backbone as PK30-
DAP80 but without interacting side chains), it is possible to highlight also the (slight)
contribution of the furan groups (probably via π-stacking interactions) in increasing the Tg
value of the original polyketone.
Figure 2.4 FT-IR spectra of PK30 modified with furfurylamine (PK30-FA) and PK30-FA conversion
with 1,2 diaminopropane (PK30-FA-DAP) at different ratios.
As expected, samples of the PK30-FA-DAP series (thus displaying the presence of both
interactions) show Tg values intermediate between the ones of the two reference compounds
(PK-FA80 and PK-DAP80) at the same total conversion value (Figure 2.5B). All DSC data
clearly demonstrate, even before cross-linking, the remarkable versatility of the employed
synthetic approach in yielding several different polymeric systems with tuneable thermal
behaviour as function of composition.
The cross-linking reactions of PK30-FA and PK30-FA-DAP with bis-maleimide were carried
out at 50 °C for 24 h. As previously reported [33], IR spectroscopy is useful to monitor the
DA cycloaddition reaction through the spectral band of C-O stretching around 1000-1300 cm-
1. Figure 2.6 shows FT-IR spectral results for the reaction of PK30-FA and PK30-FA-DAP
3000 2500 2000 1500 1000
B
C
D
A
Tra
nsm
itta
nce
(a.
u.)
Wavenumber (cm-1)
A PK30FA80
B PK30FA60DAP20
C PK30FA40DAP40
D PK30FA20DAP60
25
with bis-maleimide at different molar ratios between furan and maleimide groups. The band
centred around 1180 cm-1
corresponds with the C-O-C ether peak. Its intensity increases with
the molar ratio between furan and maleimide groups, thus testifying the occurrence of the
Diels-Alder reaction. In analogy with previous reports [33], the molar ratio between furan
and maleimide were systematically changed (Table 2.2).
Figure 2.5 Tg (as measured by DSC) of PK30 as a function of conversion percentage with
furfurylamine or diaminopropane (A) and as function of different furfurylamine / diaminopropane
ratios at 80% total conversion (B).
The mechanical behaviour of all prepared samples (as solid bars) was tested by DMTA at
three heating cycles. The corresponding Tg values (taken as peak of tan (δ) in the
corresponding DMTA curves) increase with the furan/maleimide molar ratio (for every
series), as expected on the basis of the corresponding network density. More importantly, if
0
20
40
60
80
100
120
Ratio furan / amine groups (%)
Tg (
o C
)
20 40 60 800
20
40
60
80
100
120
Tg (o
C)
Conversion (%)
PK-FA
PK-DAP
26
one compares materials with the same total carbonyl conversion but displaying the presence
of different groups along the backbone (e.g. PK30-FA80, PK30-FA60-DAP20, PK30-FA40-
DAP40, PK30-FA20-DAP60), it is clear that the corresponding Tg increases with the relative
amount of amino groups (i.e. hydrogen bonding interactions) with respect to the furan ones
(i.e. DA adduct).
Figure 2.6 FT-IR spectra of PK30 functionalized with furfurylamine (PK30-FA80) (A) and 1,2
diaminopropane (PK30-FA20DAP60) (B) cross-linked at different FA/Ma ratios. Bands normalized at
2887 cm-1
(C-H stretch).
The trend is not linear with composition (or at least it does not seem to follow a linear
pattern) since hydrogen bonding between amino groups and the carbonyls on the bis-
maleimide compound might also take place and thus decrease the amount of available –NH2
groups. Moreover, the recovery of mechanical properties (modulus in this case) seems to be
27
almost quantitative for both PK30-FA (Figure 2.7A) and PK30-FA-DAP (Figure 2.7B, for
brevity only one sample is shown).
Table 2.2 Tg values of functionalized polyketones 30 (PK30-FA and PK30-FA-DAP) as a function of
cross-link density between furan and maleimide grups.
Experiments Ratio Fa / Ma Tg ( oC ) ∆Tg
a
PK30-FA80 1 / 0.25 90 0
PK30-FA80 1 / 0.5 111 0
PK30-FA80 1 / 0.75 130 0
PK30-FA80 1 / 1 140 0
PK30-FA60-DAP20 1 / 0.25 95 0
PK30-FA60-DAP20 1 / 0.5 125 8
PK30-FA60-DAP20 1 / 0.75 135 11
PK30-FA60-DAP20 1 / 1 161 21
PK30-FA40-DAP40 1 / 0.25 125 12
PK30-FA40-DAP40 1 / 0.5 n.d. n.d.
PK30-FA40-DAP40 1 / 0.75 159 21
PK30-FA40-DAP40 1 / 1 183 32
PK30-FA20-DAP60 1 / 0.25 129 n.d.
PK30-FA20-DAP60 1 / 0.5 126 13
PK30-FA20-DAP60 1 / 0.75 155 20
PK30-FA20-DAP60 1 / 1 185 40
aTg difference between the 1
o and 3
o heating cycle. Data
obtained by DMTA measurements.
However, samples displaying the presence of both interactions (PK30-FA-DAP) show a
systematic increase in the softening temperature (see Figure 2.7B for an example) upon
thermal cycling. This phenomenon is systematic and can result (Table 2.2) in differences up
to 40 °C between the Tg measured during the last (third) and first DMTA cycle.
Although this does not seem to affect the mechanical properties (modulus) and their recovery
in a relevant way, the reasons for such behaviour were further investigated. The increase in
the softening temperature could suggest an inter-polymeric conformational re-arrangement or
a side reaction between the unreacted functional groups as an effect of temperature exposure.
To clarify this point, every sample (after the three heating cycles) was analysed by FTIR.
28
Figure 2.7 Thermal behaviour of PK30-FA80 / Ma (1/1 ratio) tested by 4 DMTA Cycles (A). PK30-
FA20-DAP60 / Ma (1/1 ratio) tested by 3 DMTA cycles (B).
The rather complicated absorption patterns did not allow a direct analysis and a
deconvolution procedure was thus applied. In particular (Figure 2.8A) the absorption in the
carbonyl region was deconvoluted yielding, as first result, the presence of an imine (Schiff
base) absorption [45] at around 1640 cm-1
. This suggests the possible reaction of the free
amino groups with unreacted carbonyls along the polymer backbone, which should be faster
and more prominent with increasing amount of amino groups along the backbone. This can
be clarified when plotting (Figure 2.8B) the ratio between the peak area at 1640 cm-1
(imine
group) with the one at 1703 cm-1
(carbonyl group) as function of the amino group intake. It is
clear (not shown for brevity) how this ratio does not change when amino groups are not
present in the system, thus proving the lack of any side reactions. On the other hand, if amino
29
groups are present in the systems such ratio significantly increases (Figure 2.8B) after the
DMTA cycles (with respect to the original material), with the magnitude of the difference
also increasing as function of the DAP intake.
Figure 2.8 (A) FT-IR deconvolution spectra of PK30-FA40-DAP40 cross-linked with bis-maleimide
evaluated after three DMTA cycles (R2 = 0.98). (B) Ratio between imine region band (1640 cm
-1) and
carbonyl band (1703 cm-1
) of PK30-FA-DAP/bis-maleimide (before and after 3 DMTA cycles) as a
function of DAP moieties intake. Data obtained from deconvolution of FT-IR spectra.
The presented data clearly show that the presence of amino groups is, for this particular
system, not a full advantage. Despite the clear increase in softening point and the still almost
quantitative recovery of elastic modulus during the thermal cycles, the formation of
permanent (i.e. not reversible in the solid state) cross-linking points through imine formation
30
might seriously hinder full reworkability of the prepared materials (at least after more than
three heating cycles). In this context, the use of other hydrogen bonding groups, not reactive
towards carbonyl ones, might represent a more viable option and is currently under
investigation.
2.4 Conclusions
A new class of polymeric materials, bearing Diels-Alder as well as hydrogen bonding active
groups along the backbone, has been successfully synthesized by chemical modification of
aliphatic polyketones. The employed synthetic strategy (Paal-Knorr reaction with primary
amines) allows easy preparation of these materials with a series of general advantages related
to this kind of polyketone modification. The thermal properties of the polymers, even before
cross-linking, testify the relevant contribution of hydrogen bonding interactions to the final
glass transition temperature values. Furthermore, the prepared polymers, after cross-linking
with a bis-maleimide, clearly showed improved Tg values with respect to the respective
counterparts containing only furan groups. From an applicative point of view, this
significantly extends the upper limit of available Tg values for these materials from 100 °C
(as reported in previous works) up to 185 °C without any significant effect (or in any case
very little) on the recovery of mechanical properties (i.e. modulus) after three thermal cycles.
However, a significant shift in the Tg values (as measured by DMTA) can be observed during
the thermal cycles. This has been preliminarily attributed to the side reaction of the free
amine groups with the unreacted carbonyls along the polymeric backbone. Such unwanted
feature renders the proposed system not entirely optimal, although significantly better than
previously reported ones, in terms of end-product reworkability. The incorporation of
hydrogen bonding groups not able to react with the free carbonyl groups is currently under
investigation, as it might pave the way towards more efficient systems.
2.5 References
1. Al-Maaded M, Madi NK, Kahraman R, Hodzic A, Ozerkan N. An Overview of Solid
Waste Management and Plastic Recycling in Qatar. Journal of Polymers and the Environment
2012;20(1):186-94.
2. Cossu R, Lai T, Pivnenko K. Waste washing pre-treatment of municipal and special waste.
J.Hazard.Mater. 2012;207:65-72.
3. Goodship V. Introduction to Plastics Recycling. 2nd ed. Shrewsbury, UK: Smithers Rapra
Technology Limited; 2007.
4. Jakubowicz I, Enebro J. Effects of reprocessing of oxobiodegradable and non-degradable
polyethylene on the durability of recycled materials. Polym.Degrad.Stab. 2012;97(3):316-21.
31
5. Muthu SS, Li Y, Hu JY, Mok PY, Ding X. Eco-Impact of Plastic and Paper Shopping
Bags. Journal of Engineered Fibers and Fabrics 2012;7(1):26-37.
6. Rajendran S, Scelsi L, Hodzic A, Soutis C, Al-Maadeed MA. Environmental impact
assessment of composites containing recycled plastics. Resources Conservation and
Recycling 2012;60:131-9.
7. Gibbs BF, Mulligan CN. Styrene toxicity: An ecotoxicological assessment.
Ecotoxicol.Environ.Saf. 1997;38(3):181-94.
8. Andrady AL. Microplastics in the marine environment. Mar. Pollut. Bull.
2011;62(8):1596-605.
9. Oehlmann J, Oetken M, Schulte-Oehlmann U. A critical evaluation of the environmental
risk assessment for plasticizers in the freshwater environment in Europe, with special
emphasis on bisphenol A and endocrine disruption. Environ.Res. 2008;108(2):140-9.
10. Flint S, Markle T, Thompson S, Wallace E. Bisphenol A exposure, effects, and policy: A
wildlife perspective. J.Environ.Manage. 2012;104:19-34.
11. Rajendran S, Hodzic A, Soutis C, Al-Maadeed MA. Review of life cycle assessment on
polyolefins and related materials. Plastics Rubber and Composites 2012;41(4-5):159-68.
12. Tsai W. Human health risk on environmental exposure to bisphenol-A: A review.
J.Environ.Sci.Health Pt.C-Environ.Carcinog.Ecotoxicol.Rev. 2006;24(2):225.
13. Kistler SS. Thermoplastic behaviour of linear and 3-dimensional polymers. J.Appl.Phys.
1940;11:769-78.
14. Flory PJ. Constitution of 3-dimensional polymers and the theory of gelation.
J.Phys.Chem. 1942;46:132-40.
15. Flory PJ, Rehner JJ. Statistical mechanics of cross-linked polymer networks II. swelling.
The Journal of Chemical Physics 1943;11(11):521-6.
16. Stockmayer WH. Theory of molecular-size distribution and gel formation in branched-
chain polymers. J.Chem.Phys. 1943;11:45-55.
17. Stockmayer WH. Theory of molecular-size distribution and gel formation in branched
polymers. II. General cross linking. J.Chem.Phys. 1944;12:125-31.
18. Kourtides DA. Thermochemical and Flammability Properties of some Thermoplastic and
Thermoset Polymers - Review. Polym.Plast.Technol.Eng. 1978;11(2):159-98.
19. Roller MB. Thermoset and Coatings Technology - Challenge of Interdisciplinary
Chemistry. Polym.Eng.Sci. 1979;19(10):692-8.
20. Berglund LA, Kenny JM. Processing Science for High-Performance Thermoset
Composites. SAMPE J. 1991;27(2):27-37.
21. Gascons M, Blanco N, Matthys K. Evolution of manufacturing processes for fiber-
reinforced thermoset tanks, vessels, and silos: a review. Iie Transactions 2012;44(6):476-89.
22. Toncelli C, De Reus D, Broekhuis AA, Picchioni F. Thermoreversibility in Polymeric
Systems. In: Amendola M, Meneghetti V, editors. Self-Healing at the NanoscaleNew York:
CRC Press; 2011, p. 199.
23. Cho SH, Andersson HM, White SR, Sottos NR, Braun PV. Polydimethylsiloxane-based
self-healing materials. Adv Mater 2006;18(8):997-1000.
24. Liu X, Lee JK, Yoon SH, Kessler MR. Characterization of diene monomers as healing
agents for autonomic damage repair. J Appl Polym Sci 2006;101(3):1266-72.
25. Pang JWC, Bond IP. A hollow fibre reinforced polymer composite encompassing self-
healing and enhanced damage visibility. Composites Sci.Technol. 2005;65(11-12):1791-9.
26. Pang JWC, Bond IP. 'Bleeding composites' - damage detection and self-repair using a
biomimetic approach. Composites Part A-Applied Science and Manufacturing
2005;36(2):183-8.
32
27. White S, Sottos N, Geubelle P, Moore J, Kessler M, Sriram S, et al. Autonomic healing of
polymer composites. Nature 2001;409(6822):794-7.
28. Ax J, Wenz G. Thermoreversible Networks by Diels-Alder Reaction of Cellulose
Furoates With Bismaleimides. Macromolecular Chemistry and Physics 2012;213(2):182-6.
29. Chen X, Dam M, Ono K, Mal A, Shen H, Nutt S, et al. A thermally re-mendable cross-
linked polymeric material. Science 2002 MAR 1;295(5560):1698-702.
30. Craven J, Wilmington C, inventors. AnonymousCroos-linked thermally reversible
polymers produced from condensation polymers with pendant furan groups cross-linked with
maleimides. . 1969 .
31. Goiti E, Huglin M, Rego J. Thermal breakdown by the retro Diels-Alder reaction of
crosslinking in poly[styrene-co-(furfuryl methacrylate)]. Macromolecular Rapid
Communications 2003;24(11):692-6.
32. Toncelli C, De Reus DC, Picchioni F, Broekhuis AA. Properties of Reversible Diels-
Alder Furan/Maleimide Polymer Networks as Function of Crosslink Density.
Macromolecular Chemistry and Physics 2012;213(2):157-65.
33. Zhang Y, Broekhuis AA, Picchioni F. Thermally Self-Healing Polymeric Materials: The
Next Step to Recycling Thermoset Polymers? Macromolecules 2009;42(6):1906-12.
34. Chang JM, Aklonis JJ. A Photothermal Reversibly Crosslinkable Polymer System.
Journal of Polymer Science Part C-Polymer Letters 1983;21(12):999-1004.
35. Burnworth M, Tang L, Kumpfer JR, Duncan AJ, Beyer FL, Fiore GL, et al. Optically
healable supramolecular polymers. Nature 2011;472(7343):334-8.
36. Zhang Q, Qu D, Wu J, Ma X, Wang Q, Tian H. A Dual-Modality Photoswitchable
Supramolecular Polymer. Langmuir 2013;29(17):5345-50.
37. Gandini A. The furan/maleimide Diels-Alder reaction: A versatile click-unclick tool in
macromolecular synthesis. Progress in Polymer Science 2013;38(1):1-29.
38. Liu Y, Chuo T. Self-healing polymers based on thermally reversible Diels-Alder
chemistry. Polymer Chemistry 2013;4(7):2194-205.
39. Zhang Y, Broekhuis AA, Stuart MCA, Picchioni F. Polymeric Amines by chemical
modifications of alternating aliphatic polyketones. J Appl Polym Sci 2008;107(1):262-71.
40. Zhang Y, Broekhuis AA, Stuart MCA, Landaluce TF, Fausti D, Rudolf P, et al. Cross-
linking of multiwalled carbon nanotubes with polymeric amines. Macromolecules
2008;41(16):6141-6.
41. Drent E, Keijsper J, inventors. AnonymousPolyketone polymer preparation with tetra
alkyl bis phosphine ligand and hydrogen. . 1993 .
42. Mul WP, Dirkzwager H, Broekhuis AA, Heeres HJ, van der Linden AJ, Orpen AG.
Highly active, recyclable catalyst for the manufacture of viscous, low molecular weight, CO-
ethene-propene-based polyketone, base component for a new class of resins. Inorg.Chim.Acta
2002;327:147-59.
43. A. I. M. Hamarneh. Novel wood adhesives from bio-based materials and polyketones.
Netherlands: University of Groningen; 2010.
44. Silverstein RM, Bassler GC, Morrill TC. Spectrometric identification of organic
compounds. Canada: John Wiley and sons, Inc.; 1991.
45. Uribe-Romo FJ, Hunt JR, Furukawa H, Klock C, O'Keeffe M, Yaghi OM. A Crystalline
Imine-Linked 3-D Porous Covalent Organic Framework. J.Am.Chem.Soc.
2009;131(13):4570-1.
33
3 Intrinsic self-healing thermosets through covalent and hydrogen bonding
interactions
Abstract: The intrinsic self-healing ability of polyketone (PK) chemically modified into
furan and/or OH groups containing derivatives is presented. Polymers bearing different ratios
of both functional groups were cross-linked via furan / bis-maleimide (Diels-Alder adducts)
and hydrogen bonding interactions (aliphatic and aromatic OH groups). The resulting
thermosets display tuneable softening points (peak of tan (δ)) from 90 to 137 °C as
established by DMTA. It is found that the cross-linked system containing only furan groups
shows the highest softening temperature. On the other hand, systems displaying the
combination of Diels-Alder adducts and hydrogen bonding (up to 60 mol % of -OH groups)
do not show any change in modulus between heating cycles (i.e. factually a quantitative
recovery of the mechanical behaviour). It is believed that the novelty of these tuneable
thermosets can offer significant advantages over conventional reversible covalent systems.
The synergistic reinforcement of both interactions resists multiple heating/healing cycles
without any loss of mechanical properties even for thermally healed broken samples.
Keywords: Paal-Knorr chemical modification, Diels-Alder reaction, Hydrogen bonding
interactions, tuneable self-healing thermoset, recycling.
34
3.1 Introduction
Over the last few decades, researchers working on the design and optimization of plastic
materials have reported considerable improvements (e.g. in mechanical properties of recycled
materials) often stimulated by societal driven demands in terms of sustainability,
enhancement of material quality and value-add services [1, 2]. However, despite all the
efforts, it is widely recognized that plastic recycling is still facing major problems regarding
collection, separation, cleaning, processing chemistry and flow markets for recycled products
[3-6].
As an attempt for improvements in this matter, the European Commission has categorized the
waste treatment according to the most environmentally favourable strategies. The “waste
hierarchy” stablishes as priority the prevention follows by the reuse and recycling of waste,
while the less favourable strategies are the recovery and disposal [7]. In the particular case of
plastics, suitable materials for recycling are subjected to the availability of treatment
technology and markets [8, 9]. For instance, the most used procedures for thermoplastics
recycling are the physical and chemical approaches. The physical recycling is realized by the
grinding and re-melting of thermoplastics to produce a material with equal, similar or
completely different properties compared to the original one. On the other hand, the chemical
recycling turns thermoplastics into monomers or oligomers (by the addition of chemical
additives and processing conditions) to be used as raw materials for the production of new
polymers [10].
Thermosets unlike thermoplastics, cannot be recycled by using the above mentioned
approaches (e.g. re-melted / re-shaped or chemical reduction to monomers) since they
degrade or decompose. In this context, (thermally) reversible thermoset materials represent
an accessible solution due to their remarkable ability of being mended and recycled
(according to a “cradle to cradle” approach) in order to prolong their use after their first
service life [11]. Several scientific articles have reported the ability of thermoset polymers to
be repaired in different ways. These include the incorporation of low molecular weight
additives in the form of capsules as healing agent after damage events (extrinsic self-repair)
[12-16], but also the functionalization of raw non-reworkable polymers that can heal cracks
as a consequence of external stimuli like heat [17-20] and light [21-25] (intrinsic self-repair).
The last approach relies on the chemical modification of the base polymer with functional
35
groups able to undergo a reversible reaction. Diels-Alder or hydrogen bonding active groups
are indeed often employed to prepare polymer networks with thermally reversible properties
[26, 27]. Therefore, in a fracture event, the healing process can be performed by the
combination of covalent (Diels-Alder) and or non-covalent (H-bonds) reversible interactions.
It is ideal as the intrinsic self-healing ability (i.e. bonds reconnection) of both functional
groups plays a crucial role in supporting the recovery of the network (e.g. dimensional
stability) at different levels of energy (i.e. complementary reinforcement of chemical and
physical interactions). Therefore, the healing process can be achieved under different
conditions (e.g. temperature) just by controlling the ratio between the introduced functional
groups. These advantages have recently triggered the research on thermoset systems with
tuneable chemical reversibility and mechanical properties (e.g. polymers containing
reversible [20, 26, 28, 29] and non-reversible [30-32] covalent interactions in combination
with hydrogen bonding). However, the design of this kind of materials still faces problems in
terms of lengthy, costly and cumbersome synthetic steps, thus hindering any application at
industrial scale.
We recently reported on the novel thermally self-healing properties of polymer networks
containing reversible covalent and non-covalent interactions based on aliphatic polyketones
[20, 33]. In order to prepare the target materials, we started modifying alternating aliphatic
polyketone (PK) into new polymers by using the Paal-Knorr reaction, i.e. the chemical
modification of the di-carbonyl arrangement of PK into pyrrole groups bearing furan or
amine active groups [20]. This synthetic pathway offers several advantages like high yield
under relatively mild condition (100 °C) and reaction kinetics (4 h) even without any catalyst,
with water as the only by-product. Subsequently, the modified PK was covalently cross-
linked through the Diels-Alder (DA) reaction by using bis-maleimide. The simultaneous
presence in the modified polymers of both interactions (H-bonding and DA adducts) resulted
in networks with softening points from 100 to 185 °C . Remarkably, the materials retained
quantitative modulus values after multiple heating-healing cycles. However, they also
displayed an increase in the softening point between cycles, which was ascribed to the
formation of irreversible cross-linking points via the reaction of the pendant amine groups
with the unreacted carbonyls along the backbone [20]. In order to overcome this drawback,
this work is focused on preparing tuneable reversible thermosets through the synergistic
cooperation of covalent and non-covalent interactions. These materials should then be able to
display reversible properties after several heating/healing and recycling cycles. In principle,
36
this can be achieved by using OH functional groups (instead of amino ones) as hydrogen
bonding active moieties, thus avoiding any side reaction with the unreacted carbonyls along
the backbone. In order to establish more clearly the role of the OH-groups, reference
polymers displaying the same pyrrolic backbone of the target ones were functionalized with
less reactive moieties without OH groups.
After modification, the polymers are covalently cross-linked via furan/maleimide Diels-Alder
and hydrogen bonds. The thermal and mechanical behaviour is then studied by differential
scanning calorimetry DSC and dynamic mechanical thermal analysis (DMTA) to determine
the reversibility, rework-ability and intrinsic self-healing ability of the thermoset polymers.
3.2 Experimental section
Reagents
The alternating aliphatic polyketone (PK30, MW 2687 Da) presents a total olefin content of
30% of ethylene and 70% of propylene [34, 35]. The polymer presents a 43% of carbonyl
content on the basis of the total molecular weight. Furfurylamine (Fu) was freshly distilled
(Aldrich, ≥ 99%). 3-amino-1-propanol (Ap) (Acros, The Netherland), 2-(4-
hydroxyphenyl)ethylamine (tyramine Ty) (Sigma Aldrich 99%), 2,5-hexanedione (Sigma
Aldrich 98%), butylamine (Ba) (Sigma Aldrich 99%) benzylamine (Bea) (Sigma Aldrich
99%), 1-Propanol, Milli-Q water, (1,1-(methylenedi-4,1-phenylene) bis-maleimide (b-Ma)
(Sigma Aldrich 95%), chloroform (CHCl3 Laboratory-Scan, 99.5%), were purchased and
used as received. Deuterated dimethylsulfoxide (DMSO-d6, Sigma Aldrich, ≥99 atom%) was
used as solvent for 1H-NMR measurements.
Model component reaction
Model reactions between stoichiometric amounts of 2,5-hexanedione (8,7 mmol) with either
furfurylamine (previously reported by our group [36]), 3-amino-1-propanol (0.65 g) or
tyramine (1.2 g) were carried out in order to identify the presence of any side product. The
reactants were poured into a flask provided with a reflux condenser and transferred to the
microwave apparatus. The reactions were performed at 100 °C (200 watts) for two hours in
bulk in the case of 2,5-hexanedione with 3-amino-1-propanol and using 1-propanol as solvent
(10 wt.%) in the case of 2,5-hexanedione with tyramine. Thereafter, the organic solvent was
37
removed in a vacuum oven at 50 ᵒC for 24 h and the by-product (H2O) was removed in a
freeze dryer for 72 hours.
Polyketone modification
The chemical modification of PK30 was performed by two different heating methods (1
conventional oil bath and 2 microwave oven). In both methods, different molar ratios
between 1,4-dicarbonyl of PK30 and amine compounds were used. PK30-Fu and PK30-Fu-
Ap were prepared in a conventional oil bath as well as the reference polymers PK30-Ba and
PK30-Bea, while PK-Fu and PK-Fu-Ty by using the microwave oven.
In the case of 1, the reactions were carried out in bulk in a sealed 250 mL round-bottomed
glass reactor with a reflux condenser, a U-type anchor impeller, and an oil bath for heating.
After 60 g of PK30 were preheated to a liquid state at 100 °C, the amine compounds
furfurylamine and/or 3-amino-1-propanol (butylamine or benzylamine for reference
polymers) were added dropwise to the reactor during the first 20 min (mmol equivalent to 20
– 80 % aimed conversion). The stirring speed was set at a constant value of 600 rpm and the
reaction time was set at 4 h.
Procedure number 2 is employed to reduce the reaction time for the chemical modification of
PK30 into polymers bearing Fu and aromatic-OH groups [37]. In a typical experiment, a
mixture of polymer (4.0 g PK30) with furfurylamine and/or tyramine (mmol equivalent to 35
– 80 % aimed conversion) was pre-dissolved in 1-propanol (≈ 10 wt.% of polymer) and then
poured into a flask provided with a reflux and transferred to the microwave apparatus. The
microwave power was set at 200 Watts to keep a constant temperature of 100 °C for 2 hours.
After reaction, the solvent was removed using rotary evaporation. Finally, the products were
dried under vacuum at 50 °C for 24 h.
In both procedures, the resulting polymers were frozen with liquid nitrogen, crushed to
powdery samples and washed three times with deionized Milli-Q water to remove unreacted
amine compounds, if any. After filtering and freeze drying, light-brown polymers were
obtained as final products. The carbonyl conversion (Cco) of PK30 can be calculated by:
𝐶𝑐𝑜 =𝑦
𝑦+𝑥 ∙ 100 (1)
38
where x and y represent the moles of di-ketone and pyrrolic units after conversion,
respectively. y can be calculated as follows:
𝑦 =𝑁
𝐴𝑚(𝑁) (2)
Where N represents the grams of nitrogen in the product according to elemental analysis and
𝐴𝑚(𝑁) represents the atomic mass of nitrogen. x can be calculated as follows:
𝑔𝑝𝑟𝑜𝑑 = 𝑥 ∙ 𝑀𝑤 𝑝𝑘 + 𝑦 ∙ 𝑀𝑤
𝑦 (3)
where gprod represents the grams of the final product after conversion, Mw y
the molecular
weight of the pyrrolic unit (i.e. pyrrolic units bearing either Fu, OH or both functional
groups) and Mw pk
the molecular weight of the 1,4 di-ketone unit. Then we can conclude that:
𝑥 =𝑔𝑝𝑟𝑜𝑑−𝑦∙𝑀𝑤
𝑦
𝑀𝑤 𝑝𝑘 (4)
The conversion efficiency η can be defined as the ratio between the final carbonyl conversion
Cco and the targeted one according to the amount of polymer and amine compounds provided
in the feed (Cco feed
):
η =𝐶𝑐𝑜
𝐶𝑐𝑜 𝑓𝑒𝑒𝑑 ∙ 100 (5)
The Cco feed
is calculated as follow:
𝐶𝑐𝑜 𝑓𝑒𝑒𝑑
=𝑀𝑜𝑙𝑎𝑚𝑖𝑛𝑒
𝑀𝑜𝑙𝑃𝐾30 ∙ 100 (6)
with Molamine representing the moles of amine compounds and MolPK30 the moles of di-
carbonyl units in the feed.
Diels-Alder reaction
The DA reaction of PK30-Fu and PK30-Fu-OH (2.0 gr) with bis-maleimide was carried out
in a ratio 1:1 between the furan moiety and the maleimide group using chloroform as solvent
(≈ 10 wt.% of polymer) in a 100 mL round-bottomed flask equipped with a magnetic stirrer
and a reflux condenser. The reaction mixture was heated up to 50 °C for 24 h to form the DA
adducts. After reaction, the cross-linked polymers were dried at 50 °C under vacuum
overnight to remove the solvent.
39
Characterization
The elemental analysis was performed using an Euro EA elemental analyzer. 1H-NMR
spectra were recorded on a Varian Mercury Plus 400 MHz apparatus using DMSO-d6 as
solvent (residual resonance at 2.5 ppm). FT-IR spectra were recorded using a Perkin-Elmer
Spectrum 2000. The samples (pellets) were prepared by mixing potassium bromide (KBr)
with the polymers at (1.5 wt.%). The pills were then kept in an oven for 24 hours at 50 °C to
remove residual water. DSC analysis was performed on a TA-Instrument DSC 2920 under N2
atmosphere. The samples for DSC were weighed (10-17 mg) in an aluminium pan, which was
then sealed. The samples were first heated from 0 to 180 °C and then cooled to 0 °C. Four
cycles were performed at 10 °C/min scanning rates. GPC measurements were performed with
a HP1100 Hewlett-Packard to evaluate possible side reactions of the polymers during the
modification step, if any. The equipment consists of three 300 x 7.5 mm PLgel 3 µm
MIXED-E columns in series and a GBC LC 1240 IR detector. The samples were dissolved in
THF at a concentration of 1 mg/mL. THF was used as eluent at a flow rate of 1 mL/min at a
temperature of 40 °C. The calibration curve was realized using polystyrene as standard and
the data were determined using PSS WinGPC software. The samples for DMTA analysis
were prepared by compression molding of 500 mg of cross-linked PK30-Fu / Bis-maleimide
and PK30-Fu-OH / Bis-maleimide into rectangular bars of 6 mm in width, 1 mm in thickness,
and 54 mm in length at 150 °C for 30 min under a pressure of 40 bar. The samples were then
kept in an oven at 50 °C for 24 h aimed at thermal healing treatment. The DMTA analysis of
the bars was conducted on a Rheometrics scientific solid analyzer (RSA II) under an air
environment using the dual cantilever mode at an oscillation frequency of 1 Hz and a heating
rate of 3 °C/min.
3.3 Results and discussion
Model component Paal-Knorr reaction
Model reactions of 2,5-hexanedione (i.e. representative for the di-carbonyl moieties along the
PK30 backbone) with aliphatic and aromatic amines were performed. The present Paal-Knorr
reaction resulted in the formation of a pyrrole group, ultimately yielding pyrrole-furane [36]
and pyrrole-OH derivatives (see Figure 3.1 for H-NMR spectra). In Figure 3.1A it is possible
to notice the signal of protons associated to the pyrrole ring at 5.6 ppm, the presence of CH2
units at 1.6, 3.4 and 3.8 ppm and the proton signal at 4.6 corresponding to the OH group. In a
40
similar way, Figure 3.1B displays the proton signals associated to the pyrrole ring at 5.6 ppm,
the presence of CH2 units at 2.7 and 3.8 ppm, the proton signals associated to the aromatic
ring at 6.6 and 6.9 ppm and the proton signal corresponding to the OH group at 9.2. In both
cases, no relevant by-products are observed and, in particular, no evidence is found for a
possible reaction between the –OH and carbonyl groups.
Figure 3.1 1H-NMR spectra of model compound reaction between 2,5-hexanedione and 3-amino-1-
propanol A and tyramine B. The peak at 3.3 ppm can be assigned to residual water.
Paal-Knorr reaction of PK30
The aim of this work is the preparation of thermo-reversible polymer networks by combining
reversible covalent bonding (DA and r-DA sequence) using PK-furan derivatives together
with OH groups. We started with the conversion of PK into the PK-Furan-OH (i.e aliphatic or
aromatic OH derivatives) by Paal-Knorr reaction by using a mixture of the corresponding
amino compounds (Figure 3.2).
Figure 3.2 Schematic representation of polyketone PK functionalized with aliphatic and aromatic
amines by Paal-Knorr reaction.
Since the amine conversion is practically quantitative in this reaction [20, 36], this easily
allows (in principle) the preparation of materials characterized by different conversion values
6 5 4 3 2 1
ppm
9 8 7 6 5 4 3 2 1
ppm
41
of carbonyl groups. The reactions between PK30 and aliphatic / aromatic amine compounds
were carried out according to different molar ratios as expressed in the sample nomenclature
(Table 3.1). The corresponding samples are coded as PK30-Fu X1-OH X2 with xi being the
mol percentage of amine with respect to carbonyl groups in the feed while OH represents a
common symbol for 3-amine-1-propanol (Ap) or tyramine (Ty).
Table 3.1 Experimental conditions and results of the chemical modification of PK30 using different
ratios between aliphatic and aromatic amine compounds ( 𝑋𝑁𝐻2).
Oil bath 𝐗𝐍𝐇𝟐/𝐂=𝐎 𝐗𝐍𝐇𝟐
(moles)
CCO a
(%)
ηb
(%) Microwave
𝐗𝐍𝐇𝟐/𝐂=𝐎 𝐗𝐍𝐇𝟐
(moles)
CCO a
(%)
ηb
(%)
PK30-Ba80* 0.8 0.365 78 98 PK30-Bea80* 0.8 0.024 69 86
PK30-Fu20 0.2 0.091 19 99 PK30-Fu80 0.8 0.024 68 85
PK30-Fu40 0.4 0.182 39 99 PK30-Ty35 0.35 0.011 34 97
PK30-Fu60 0.6 0.274 58 98 PK30-Ty50 0.5 0.015 49 98
PK30-Fu80 0.8 0.365 76 96 PK30-Ty65 0.65 0.020 63 97
PK30-Ap20 0.2 0.091 15 75 PK30-Ty80 0.8 0.024 76 95
PK30-Ap40 0.4 0.182 36 90 PK30-Fu65-Ty15 0.8 0.024 60 75
PK30-Ap60 0.6 0.274 57 95 PK30-Fu50-Ty30 0.8 0.024 67 83
PK30-Ap80 0.8 0.365 79 99 PK30-Fu35-Ty45 0.8 0.024 73 91
PK30-Fu20-Ap60 0.8 0.365 77 96 PK30-Fu15-Ty65 0.8 0.024 73 91
PK30-Fu40-Ap40 0.8 0.365 78 97
PK30-Fu60-Ap20 0.8 0.365 78 98
a % conversion of carbonyl groups ,
b η conversion efficiency.*reference samples: PK30-Ba
(butylamine)and PK30-Bea (benzylamine) same polymer structure but not bearing active functional
Fu (furan) or OH (propanol, phenol) groups)
The obtained results (Table 3.1) include the carbonyl conversion (CCO, as measured
experimentally) and the relative efficiency for the main conversion (η). Indeed, in terms of
conversion efficiency, almost all used amines display values around 100%, thus testifying the
easiness of the reaction as well as its quantitative yield [20, 36]. Moreover, the reported data
in Table 3.1 confirm that the use of microwaves can significantly shorten the reaction time
with no detrimental effect on the conversion values [20, 36, 38, 39]. However, minor
42
detrimental effects were observed on the molecular weight of polymers modified with only
tyramine according to GPC data (Table 3.2).
Table 3.2 GPC measurements of PK30 modified with Butilamine (Ba), benzylamine (Bea), 3-amino-
1-propanol (Ap), tyramine (Ty) and furfurylamine (Fu) at a maximal 80% of conversion and different
ratios between Fu and amine compounds bearing OH groups (Ap and Ty).
Run Mn (x 10
3) Mw (x 10
3) PDI
PK30-Ba80 3.4 6.1 1.8
PK30-Bea80 2.9 6.5 2.2
PK30-Ap80 2.7 6.2 2.3
PK30-Ty35 2.8 6.7 2.4
PK30-Ty50 2.6 5.5 2.1
PK30-Ty65 2.5 4.7 1.9
PK30-Ty80 2,0 3,3 1.7
PK30-Fu80 2,8 6,5 2.3
PK30-Fu60-Ap20 3.3 6.4 1.9
PK30-Fu40-Ap40 2.9 6.1 2
PK30-Fu20-Ap60 2.8 5.8 2
PK30-Fu65-Ty15 3,0 7,0 2.3
PK30-Fu50-Ty30 3,1 7,1 2.3
PK30-Fu35-Ty45 3,1 6,5 2.1
PK30-Fu15-Ty65 2,9 5,6 1.9
Besides, slightly lower η values are obtained for PK30-Fu-Ty when the concentration of Fu is
increased, in agreement with the sensitivity of this reaction to steric hindrance [39, 40]. The
1H-NMR spectra of the polymers bearing different ratios of pendant groups are displayed in
Figure 3.3. As mentioned earlier, model compounds can be used as simple representations of
complex polymeric systems. In the case of Figure 3.3, the overlapping and broad character of
many peaks was easily overcome using the information obtained from the model compounds.
Figure 3.3A shows signals that belong to protons associated to pyrrole rings at 5.6 ppm, the
presence of CH2 units attached to the pyrrol groups in the OH moiety at 3.3 and 3.7 ppm and
the proton signal of OH groups at 4.6 ppm. In a similar way, the presence of CH2 unit
between the pyrrol and furan group is noticed at 4.9 ppm and proton signals of furan moieties
at 6.1, 6.3 and 7.5 ppm are observed. Figure 3.3b evidences the formation of the pyrrole ring
at 5.6 ppm, the presence of CH2 unit attached to the pyrrol group in the OH moiety at 3.8
43
ppm, the protons associated to the benzene ring at 6.6 and 6.9 ppm, and the protons of OH
groups at 9.2 ppm.
Figure 3.3 1H-NMR spectra of PK30 modified with different ratios between furane (Fu) and OH
groups (aliphatic-OH a and aromatic-OH b) at a maximal 80% of conversion.
The area below each signal changes according to the number of protons that belong to each
functional group whereas the intensity of the signal assigned to the pyrrole ring remains
constant. This trend well agrees with the formation of the desired structure, although a
quantitative estimation is affected by peak overlapping. Furthermore, spectral data obtained
by FT-IR analysis corroborate the 1H-NMR results. Figure 3.4A displays FT-IR spectra for
PK30-Fu-Ap at two different conversion values. We start by noticing the appearance of broad
peaks around 3400 cm-1
that belong to the OH group experiencing intermolecular hydrogen
bonding, the pyrrole ring at 3115 and 1505 cm-1
and the furan moiety at 3150, 1145 and 738
cm-1
.
8 7 6 5 4 3
ppm
10 9 8 7 6 5 4
ppm
44
Figure 3.4 FT-IR spectra of PK30 functionalized with furan / OH groups at different ratios, aliphatic
OH (A) and aromatic OH (B). Figure C displays the magnification in the C-H bending region for B
(furan peak 742 cm-1
and benzene peak 827 cm-1
). Bands normalized at 2962 cm-1
(C-H stretch).
The same holds true for PK30-Fu-Ty (Figure 3.4B) with the expected appearance of
absorptions related to the benzyl ring at 1613 and 828 cm-1
. In both cases (Figures 3.4A and
3.4B the intensity between the peaks of furan and OH groups change as a function of
conversion. Figure 3.4C displays the magnification of the PK30-Fu-Ty spectra in the C-H
3500 3000 2500 2000 1500 1000 500
Tra
nsm
itta
nce
(a.
u.)
Wavenumber (cm-1)
PK30-Fu65-Ty15
PK30-Fu15-Ty65
3500 3000 2500 2000 1500 1000 500
Tra
nsm
itta
nce
(a.
u.)
Wavenumber (cm-1)
PK30-Fu60-Ap20
PK30-Fu20-Ap60
850 825 800 775 750 725 700
Ab
sorb
ance
Wavenumber (cm-1
)
PK30-Fu80
PK30-Fu50-Ty30
PK30-Fu15-Ty65
45
bending region for polymers bearing only furan or both furan and benzyl-OH groups at
different conversion values. The figure clearly evidences the furan peak decreasing as
conversion becomes in favour of benzyl-OH groups (i.e. out-of-plane C-H bending). All
prepared polymers were characterized, before cross-linking, by DSC analysis (Figure 3.5).
Figure 3.5 Tg (measured by DSC) of PK30: as a function of different percentage of conversion with
furfurylamine (PK30Fu), 3-amino-1-propanol (PK30Ap) and tyramine (PK30Ty) (A), as a function of
different interacting side groups but same (~80%) total conversion and backbone structure (B) and as
a function of different Fu / OH ratios at 80% total conversion (C).
Polymers with only one functional group (either Fu or OH) display a monotonous increase of
Tg with the amine conversion (Figure 3.5A). This is in agreement with the increased rigidity
of the backbone by increasing the amount of pyrrole groups. Moreover, PK30 functionalized
with aromatic OH groups (PK30-Ty) systematically displays (at equal conversion values)
significantly higher Tg than those modified with furane (PK30-Fu) and aliphatic OH groups
(PK30-Ap). This behaviour is attributed to the bulkiness of tyramine (which increases the
steric hindrance so that the energy demand to activate segmental motions) and the relevant
0
10
20
30
40
50
60
70
80
90
100
Tg(o
C)
10 20 30 40 50 60 70 80
0
10
20
30
40
50
60
70
80
90
100
Tg(o
C)
Conversion (%)
30
40
50
60
70
80
90
100
Tg(o
C)
Ratio Fu/OH grups (%)
PK30-Fu-Ap
PK30-Fu-Ty
46
contribution (differences are up to 60 °C) provided by hydrogen bonding interactions. Also,
PK30-Ty displays higher Tg values than PK30-Ap possibly due to the more acidic nature of
phenyl-OH groups. This feature can be confirmed by the broad character of signals associated
to hydrogen bonding of OH groups around 3420-3390 cm-1
and considering the evident
bathochromic shift (30 cm-1
for PK30-Fu-Ty) by increasing the amount of tyramine moieties
(Figure 3.4B). Moreover, taking into account the Tg values of PK30-Ba80 (14 °C) and PK30-
Bea80 (42 °C), i.e. the polymers with no interacting side chains since modified by butylamine
and benzylamine, respectively, the contribution of hydrogen bonding is evidenced (Figure
3.5B). As expected, samples of the PK30-Fu-OH series (thus displaying the presence of both
hydrogen bonding and possibly aromatic interactions (e.g PK30-Fu-Ty)) show Tg values
intermediate between the ones of the two reference compounds (PK30-Fu80 and PK30-
OH80) at the same total conversion value (Figure 3.5C). It is interesting to note in Figure
3.5C a deviation from linearity for PK30-Fu65-Ty15. This can possibly be ascribed to the
total lower conversion of this system, compared to the others (see Table 3.1). It is worth
noting that all DSC data clearly demonstrate, even before cross-linking, the remarkable
versatility of the employed synthetic approach in yielding polymeric systems with tuneable
thermal behaviour as a function of composition.
On the other hand, it can also be speculated that Tg values may come from undesired cross-
linking reactions occurring during the polymer modifications. In order to clarify this point,
Gel Permeation Chromatography analysis was performed on the different products (Table
3.2). According to these results, it can be recognized that all systems display similar values of
the polydispersity index (PDI) and practically the same magnitude of Mn and Mw, which
proves that the different values obtained for Tg are only consequences of pyrrole formation,
inter-polymeric interactions of functional groups and steric hindrance according to the
bulkiness of each modifier as above mentioned.
The polymers were then cross-linked and de-cross-linked via a DA and r-DA sequence using
an aromatic bis-maleimide (Figure 3.6). The cross-linking reactions of PK30-Fu and PK30-
Fu-OH with bis-maleimide occurred at 50 oC for 24 h. All prepared samples were cross-
linked with bis-maleimide in a ratio 1:1 between furane and maleimide groups. In a first step,
DSC analysis was carried out in order to investigate the thermal behaviour of the prepared
samples (Figure 3.7).
47
Figure 3.6 Schematic representation of DA / r-DA sequence and molding of functionalized
polyketone PK-Fu-OH cross-linked with bis-maleimide. In red reversible covalent interactions and
blue hydrogen bonding (aliphatic or aromatic).
In the heating and cooling runs, the occurrence of the Diels-Alder reaction (exothermic
peaks) and its reverse process (endothermic peaks) are clearly visible. It is worth noting that
each sample showed practically the same thermogram curves up to 3 cycles between 0 to 180
oC, thus confirming the full system reversibility. Moreover, the endothermic transition phase
in all samples took place over a broad range of temperatures that reached values up to twenty
degrees from the beginning to the end of the r-DA (i.e. the highest peak in the endothermic
transition phase).
On the other hand, it is worth mentioning that the r-DA enthalpy (i.e. the area under the curve
associated to the r-DA peak) is related to the energy required to break all DA adducts in each
particular cross-linked system. According to this, one could assume in advance that the
system containing the higher amount of furan/maleimide adducts (e.g PK30-Fu80/Ma) should
display the higher value of enthalpy when compared to those samples displaying the
combination with Fu/b-Ma and hydrogen bonding OH groups. Indeed, the enthalpy value
decreases (Figure 3.7A= 33.66 J/g > 3.7B= 11.18 J/g > 3.7C= 8.17 J/g) with decreasing
amount of DA adducts. Notably, the almost quantitative overlap of the endothermic and
exothermic traces clearly indicates the reversibility of the systems upon thermal treatment at
least in terms of amount of DA adducts broken and formed, respectively.
48
Figure 3.7. DSC thermal cycles of cross-linked PK30-Fu80 (A), PK30-Fu40-Ap40 (B), PK30-Fu35-
Ty45 (C). Ratio Fu / Maleimide = 1. Only three samples are showed for brevity.
The thermo-mechanical behaviour of all prepared samples (as solid bars) was tested by
DMTA at four heating cycles. DMTA results demonstrated that the corresponding softening
point (taken as peak of tan (δ) in the corresponding DMTA curves) increases with the amount
of furan/maleimide adducts in systems bearing only Fu groups. This is expected on the basis
of the corresponding network density (Figure 3.8).
0 20 40 60 80 100 120 140 160 180-6
-5
-4
-3
-2
-1
0
1
2
3
Temperature (oC)
Cycle
1
2
3
0 20 40 60 80 100 120 140 160 180
-5
-4
-3
-2
-1
0
1
2
Hea
t F
low
(w
/g)
end
o u
p
Temperature (oC)
Cycle
1
2
3
0 20 40 60 80 100 120 140 160 180-6
-5
-4
-3
-2
-1
0
1
2
3
Temperature (oC)
Cycle
1
2
3
0 20 40 60 80 100 120 140 160 180-4
-3
-2
-1
0
1
Temperature (oC)
Cycle
1
2
3
0 20 40 60 80 100 120 140 160 180-4
-3
-2
-1
0
1
Temperature (oC)
Cycle
1
2
3
0 20 40 60 80 100 120 140 160 180
-5
-4
-3
-2
-1
0
1
2
Hea
t F
low
(w
/g)
end
o u
p
Temperature (oC)
Cycle
1
2
3
0 20 40 60 80 100 120 140 160 180-6
-5
-4
-3
-2
-1
0
1
2
3
Temperature (oC)
Cycle
1
2
3
0 20 40 60 80 100 120 140 160 180-4
-3
-2
-1
0
1
Temperature (oC)
Cycle
1
2
3
0 20 40 60 80 100 120 140 160 180
-5
-4
-3
-2
-1
0
1
2
Hea
t F
low
(w
/g)
endo u
p
Temperature (oC)
Cycle
1
2
3
0 20 40 60 80 100 120 140 160 180
-5
-4
-3
-2
-1
0
1
2
Hea
t F
low
(w
/g)
end
o u
pTemperature (
oC)
Cycle
1
2
3
0 20 40 60 80 100 120 140 160 180-6
-5
-4
-3
-2
-1
0
1
2
3
Temperature (oC)
Cycle
1
2
3
0 20 40 60 80 100 120 140 160 180-4
-3
-2
-1
0
1
Temperature (oC)
Cycle
1
2
3
A
B
C
49
Figure 3.8 Softening points (tan δ) tested by DMTA of PK30-Fu at different grade of conversions and
PK30-Fu-OH systems at different ratios of functional groups cross-linked with bis-maleimide. Ratio
Fu/Maleimide 1:1.
In systems containing both Fu and OH functional groups (i.e. at the same 80% of carbonyl
conversion along the backbone), the corresponding softening point decreases with the relative
amount of OH groups (i.e. hydrogen bonding interactions) with respect to the furan ones (i.e.
DA adduct). The trend is not linear with composition (or at least it does not seem to follow a
linear pattern) since hydrogen bonding between OH groups and the carbonyl ones on the bis-
maleimide compound might also take place and thus decrease the amount of available OH
groups. In addition, the recovery of mechanical properties (modulus in this case) seems to be
fully quantitative for both PK30-Fu (Figure 3.9A) and PK30-Fu-OH (Figure 3.9B and 3.9C;
for brevity only three samples are shown).
0
20
40
60
80
100
120
140
So
ften
ing
po
int (
oC
PK30-Fu/Ma PK30-Fu-Ap/Ma PK30-Fu-Ty/Ma
50
Figure 3.9 Dynamic thermo-mechanical behaviour of cross-linked PK30-Fu80 (A). PK30-Fu20-Ap60
(B). PK30-Fu35-Ty45 (C). E’: storage modulus, E”: loss modulus and Tan δ: softening point
(dumping factor). Ratio Fu / Maleimide = 1.
Despite the relatively high softening point and the fully quantitative recovery of modulus
during the thermal cycles and recycling (grained samples), the formation of reversible cross-
30 40 50 60 70 80 90 100 110 120 130 140 150
1E7
1E8
1E9
Cycles
1
2
3
4 Grained
Temperature (oC)
E' E
'' (P
a)
E'
E''
0,0
0,1
0,2
0,3
0,4
tan ()
30 40 50 60 70 80 90 100 110 120 130 140 150
1E7
1E8
1E9
Cycles
1
2
3
Grained
Temperature (oC)
E' E
'' (P
a)
0,0
0,1
0,2
0,3
0,4
tan ()
E'
E''
30 40 50 60 70 80 90 100 110 120 130 140 150
1E7
1E8
1E9
Cycles
1
2
3
4 Grained
Temperature (oC)
E' E
'' (P
a)
E'
E''
0,0
0,1
0,2
tan ()
51
linking through DA reversible adducts seems to be a better option for the creation of
materials with higher damping capabilities. However, the combination with hydrogen
bonding discloses very interesting and promising pathways towards the design of tuneable (in
terms of thermal properties) reversible thermosets. Actually, considerable compensation on
the lack of DA cross-links is observed since no detrimental values of modulus are observed
by the presence of hydrogen bonds. In fact, by comparing the three different systems in
Figure 3.9, it is possibly to see that the elastic and loss modulus are increased by the
contribution of hydrogen bonding. It is worth mentioning that the ratio E” / E’ = Tan δ (right
axes in Figure 3.9) reach values of 3.9A = 0.15 < 3.9C = 0.28 < 3.9B = 0.32 indicating the
formation of more elastic materials by decreasing the Diels-Alder crosslink density,
respectively.
In these particular systems (PK-Fu-OH), the addition of OH groups is crucially from two
different points of view:
- from a scientific perspective it allows comparing polymers with the same backbone
structure (vide supra), which in turn allows pinpointing the exact influence of
hydrogen bonding on the thermal and mechanical behaviour;
- from an applicative point of view, it provides an easy and straightforward way to
modulate the softening point of the final product while decreasing the chance of side
reactions due to the fact that the carbonyl conversion is factually close to its
theoretical maximum of 80% [38]. Moreover, the presence of OH groups increases
the internal motion (friction) and relaxation of the polymeric chains during the applied
stress. This certainly increases the intrinsic self-healing properties of the materials. It
can be noticed by the fact that elastic and loss modulus remains practically as the
original in continuous DMTA measurements (Figure 3.9) due to the faster recovery of
the hydrogen bonds.
3.4 Conclusions
This work illustrates the facile synthesis of intrinsic thermally self-healing polymeric material
based on the chemical modification of aliphatic polyketones by Paal-Knorr reaction. The
modification of the polymers was proved to be feasible using conventional oil bath and
microwaves heating to achieve similar end-products. Besides the easiness of the modification
procedure, which constitutes a relevant novelty in the open literature, the Paal–Knorr reaction
52
with aliphatic and aromatic amine compounds provides a series of general advantages. In
particular, it allows the preparation of a series of compounds displaying the same backbone
structure but a (systematic) variation in the amount of hydrogen bonding and Diels-Alder
active groups. From a scientific perspective this allows to pinpoint exactly the influence of
both kinds of interactions on the thermal and mechanical behaviour. DSC traces indicated
that furan / OH-functionalized polyketones are capable to be repeatedly cross-linked and de-
cross-linked with bis-maleimide by only using heat as external stimulus. Furthermore,
DMTA measurements testify the relevant contribution of Diels-Alder covalent bonding to the
final softening point temperature while hydrogen bonding interactions seem to contribute to a
faster and quantitative recovery of modulus in these systems.
3.5 References
1. Rajendran S, Hodzic A, Soutis C, Al-Maadeed MA. Review of life cycle assessment on
polyolefins and related materials. Plastics Rubber and Composites 2012;41(4-5):159-168.
2. Nishida H. Development of materials and technologies for control of polymer recycling.
Polymer Journal 2011;43(5):435-447.
3. Henshaw J, Han W, Owens A. An overview of recycling issues for composite materials. J
Thermoplast Compos Mater 1996;9(1):4-20.
4. Yang Y, Boom R, Irion B, van Heerden D, Kuiper P, de Wit H. Recycling of composite
materials. Chem Eng Process 2012;51:53-68.
5. Pimenta S, Pinho ST. Recycling carbon fibre reinforced polymers for structural
applications: Technology review and market outlook. Waste Manage 2011;31(2):378-392.
6. Pickering S. Recycling technologies for thermoset composite materials - current status.
Composites Part A-Applied Science and Manufacturing 2006;37(8):1206-1215.
7. EUROPEAN COMMISSION. Report from the commission to the European Parliament,
the Council, the European Economic and Social Committee and the Committee of the
regions, on the Thematic Strategy on the Prevention and Recycling of Waste 2011;COM 13
final(SEC 70 final).
8. Goodship V. Introduction to Plastics Recycling. Shrewsbury, UK: Smithers Rapra
Technology Limited, 2007.
9. Andreoni V, Saveyn HGM, Eder P. Polyethylene recycling: Waste policy scenario analysis
for the EU-27. J Environ Manage 2015;158:103-110.
10. Koo HJ, Chang GS, Kim SH, Hahm WG, Park SY. Effects of recycling processes on
physical, mechanical and degradation properties of PET yarns. Fibers and Polymers
2013;14(12):2083-2087.
11. Zhang MQ, Rong MZ. Intrinsic self-healing of covalent polymers through bond
reconnection towards strength restoration. Polymer Chemistry 2013;4(18):4878-4884.
12. Cho SH, Andersson HM, White SR, Sottos NR, Braun PV. Polydimethylsiloxane-based
self-healing materials. Adv Mater 2006;18(8):997-1000.
13. Liu X, Lee JK, Yoon SH, Kessler MR. Characterization of diene monomers as healing
agents for autonomic damage repair. J Appl Polym Sci 2006;101(3):1266-1272.
14. Pang JWC, Bond IP. A hollow fibre reinforced polymer composite encompassing self-
healing and enhanced damage visibility. Composites Sci Technol 2005;65(11-12):1791-1799.
53
15. Pang JWC, Bond IP. 'Bleeding composites' - damage detection and self-repair using a
biomimetic approach. Composites Part A-Applied Science and Manufacturing
2005;36(2):183-188.
16. White S, Sottos N, Geubelle P, Moore J, Kessler M, Sriram S, Brown E, Viswanathan S.
Autonomic healing of polymer composites. Nature 2001;409(6822):794-797.
17. Long KN. The mechanics of network polymers with thermally reversible linkages. J
Mech Phys Solids 2014;63:386-411.
18. Nielsen C, Weizman H, Nemat-Nasser S. Thermally reversible cross-links in a healable
polymer: Estimating the quantity, rate of formation, and effect on viscosity. Polymer
2014;55(2):632-641.
19. Zhang R, Yu S, Chen S, Wu Q, Chen T, Sun P, Li B, Ding D. Reversible Cross-Linking,
Microdomain Structure, and Heterogeneous Dynamics in Thermally Reversible Cross-Linked
Polyurethane as Revealed by Solid-State NMR. J Phys Chem B 2014;118(4):1126-1137.
20. Araya-Hermosilla R, Broekhuis AA, Picchioni F. Reversible polymer networks
containing covalent and hydrogen bonding interactions. European Polymer Journal
2014;50:127-134.
21. Song Y, Chung C. Repeatable self-healing of a microcapsule-type protective coating.
Polymer Chemistry 2013;4(18):4940-4947.
22. Wang Z, Urban MW. Facile UV-healable polyethylenimine-copper (C2H5N-Cu)
supramolecular polymer networks. Polymer Chemistry 2013;4(18):4897-4901.
23. Chang JM, Aklonis JJ. A Photothermal Reversibly Crosslinkable Polymer System.
Journal of Polymer Science Part C-Polymer Letters 1983;21(12):999-1004.
24. Zhang Q, Qu D, Wu J, Ma X, Wang Q, Tian H. A Dual-Modality Photoswitchable
Supramolecular Polymer. Langmuir 2013;29(17):5345-5350.
25. Burnworth M, Tang L, Kumpfer JR, Duncan AJ, Beyer FL, Fiore GL, Rowan SJ, Weder
C. Optically healable supramolecular polymers. Nature 2011;472(7343):334-338.
26. Ishida K, Weibel V, Yoshie N. Substituent effect on structure and physical properties of
semicrystalline Diels-Alder network polymers. Polymer 2011;52(13):2877-2882.
27. Döhler D, Michael P, Binder W. Principles of Self-Healing Polymers. In: Binder WH,
editor. Weinheim, Germany: Wiley-VCH Verlag GmbH and Co., 2013. p. 66.
28. Skene WG, Lehn JMP. Dynamers: Polyacylhydrazone reversible covalent polymers,
component exchange, and constitutional diversity. Proc Natl Acad Sci U S A
2004;101(22):8270-8275.
29. Skene WG, Lehn JM. Dynamers: Polyacylhydrazone reversible covalent polymers.
Abstracts of Papers of the American Chemical Society 2004;228:U473-U473.
30. Khor SP, Varley RJ, Shen SZ, Yuan Q. Thermo-reversible healing in a crosslinked
polymer network containing covalent and thermo-reversible bonds. J Appl Polym Sci
2013;128(6):3743-3750.
31. Ware T, Hearon K, Lonnecker A, Wooley KL, Maitland DJ, Voit W. Triple-Shape
Memory Polymers Based on Self-Complementary Hydrogen Bonding. Macromolecules
2012;45(2):1062-1069.
32. Chen Y, Guan Z. Self-assembly of core-shell nanoparticles for self-healing materials.
Polymer Chemistry 2013;4(18):4885-4889.
33. Zhang Y, Broekhuis AA, Picchioni F. Thermally Self-Healing Polymeric Materials: The
Next Step to Recycling Thermoset Polymers?. Macromolecules 2009;42:1906-1912.
34. Drent E, Keijsper J. Polyketone polymer preparation with tetra alkyl bis phosphine ligand
and hydrogen 1993;US 5225523.
35. Mul WP, Dirkzwager H, Broekhuis AA, Heeres HJ, van der Linden AJ, Orpen AG.
Highly active, recyclable catalyst for the manufacture of viscous, low molecular weight, CO-
54
ethene-propene-based polyketone, base component for a new class of resins. Inorg Chim Acta
2002;327:147-159.
36. Zhang Y, Broekhuis AA, Picchioni F. Thermally Self-Healing Polymeric Materials: The
Next Step to Recycling Thermoset Polymers?. Macromolecules 2009;42(6):1906-1912.
37. Hamarneh AIM. NovelWood Adhesives from Bio-based Materials and Polyketones 2010.
38. Toncelli C, De Reus DC, Picchioni F, Broekhuis AA. Properties of Reversible Diels-
Alder Furan/Maleimide Polymer Networks as Function of Crosslink Density.
Macromolecular Chemistry and Physics 2012;213(2):157-165.
39. Zhang Y, Broekhuis AA, Stuart MCA, Picchioni F. Polymeric Amines by chemical
modifications of alternating aliphatic polyketones. J Appl Polym Sci 2008;107(1):262-271.
40. Hamarneh AIM. Novel wood adhesives from bio-based materials and polyketones
2010;2:22.
55
4 Thermally reversible rubber-toughened thermoset networks via Diels-Alder
chemistry
Abstract. In this work we present a reversible and toughened thermoset system based on the
covalent incorporation of a furan-functionalized ethylene-propylene rubber (EPM-Fu) into a
thermoset furan-functionalized polyketone (PK-Fu) via Diels-Alder (DA) reversible cross-
linking with bis-maleimide (b-MA). FT-IR and DSC analyses proved the reversible
interaction between PK-Fu and EPM-Fu with b-Ma via DA and r-DA sequence. Likewise,
thermo-mechanical experiments (DMTA) indicated the reworkability of the material with no
evident differences in elastic and loss modulus after several heating cycles and recycling
steps. Moreover, a considerable increase in the softening point (tan δ) was also found for the
toughened system containing 12 wt. % of EPM-Fu (137 °C for the original thermoset and 155
°C for the toughened one). A two-fold increase in Izod impact strength compared to the
original thermoset (up to 27 J/m) was also recorded for the toughened system. Overall, this
approach clearly indicates that fully thermally reversible and toughened thermosets can be
realized starting from mixtures of furan-functionalized polyketone and EPM rubber, cross-
linked via reversible Diels-Alder chemistry.
Keywords: rubber toughening, reversible thermoset, Diels-Alder reaction, recycling.
56
4.1 Introduction
Thermosetting resins are polymers that upon heat or light radiation convert into insoluble
polymer networks being cross-linked by covalent chemical bonds. Due to this network,
thermosets offer many superior properties compared to thermoplastics such as mechanical
strength, dimensional stability at elevated temperature and solvent resistance [1]. These
features confer them the ability to be widely used as composites, adhesives and coatings.
Unlike thermoplastics, thermoset cannot be re-melted or re-shaped, since they degrade or
decompose upon heating. These features limit their recyclability [2] so that at the end of their
service life, they are simply grinded to produce cheap reinforcement fillers or are thermally
processed (pyrolysis, incineration) to recover the fiber content or just energy [3-5]. A more
sustainable strategy for recycling is to re-design thermosets by replacing classical chemical
cross-links with reversible covalent bonds [2] that can be repeatedly broken and re-connected
by heating. These thermoreversible networks combine many mechanical properties of
thermosets with the processability of thermoplastics [6], but with the added value of intrinsic
self-healing properties [7]. Among several reversible reactions, the Diels-Alder (DA) [4π+2π]
cycloaddition between furan and maleimide moieties is one of the most exploited in
thermoreversible cross-linking [8]. First, DA is an equilibrium reaction influenced by
temperature: covalent furan/maleimide adducts are formed at about 50 °C (cross-linking) and
broken (de-cross-linking via retro DA) at about 120 °C. Furthermore, the cross-linking – de-
cross-linking process can be repeated many times with negligible degradation [9] in the range
of 50 °C – 150 °C. Finally, the strong dienic character of the furan ring and the high
reactivity of the maleimide as a dienophile, ensure fast kinetics and high yields [10]. Most of
the studied thermoreversible networks are based on DA reaction of polymers bearing pendant
furan and/or maleimide groups, but their syntheses are often too expensive or complex to
permit an industrial scale up. In contrast, a system based on aliphatic polyketones with
pending furan groups and cross-linked with 1,1’-(methylenedi-4,1-phenylene)bis-maleimide
was proposed as a promising low cost and efficient alternative [11]. Alternating aliphatic
polyketones, obtained by copolymerization of carbon monoxide, ethylene and propylene were
modified through the Paal-Knorr reaction to introduce furan groups directly attached to the
backbone chain. The modification reaction proceeds in bulk with high yields and fast
kinetics, producing water as only by-product. The thermoset obtained upon cross-linking with
bis-maleimide was re-healed up to seven times, displaying quantitatively retained mechanical
57
properties [11]. Furthermore, thermal and mechanical properties could be modulated as a
function of the degree of furan functionalization and the furan/maleimide molar ratio [12].
Besides the poor recyclability of thermosets currently on the market, they generally show a
high brittleness compared to thermoplastics, due to their inadequacy in dissipating energy
under mechanical stress. This causes poor resistance to crack formations as well as low
fracture toughness and impact strength [1]. This point constitutes a major drawback in the
application of thermosets, so that many systems were developed aimed at enhancing
toughness in epoxy resins, being these materials widely used for structural applications.
Those systems are essentially based on the incorporation of either rigid or soft toughening
agents in the epoxy matrix as a separate phase [13]. These modifiers can be inorganic fillers
as silica, glass beads [14, 15], engineered thermoplastics as polyphenylene oxide [16, 17],
rubbery core-shell particles [18-20] or liquid reactive rubbers such as carboxyl-terminated
(CTBN) and amine-terminated butadiene-acrylonitrile copolymers (ATBN) [21-23]. CTBN
and ATBN are the most investigated and widely used on commercial scale since they
improve toughness of the epoxy matrix without significantly affecting its bulk properties such
modulus or Tg [1]. As a common procedure, the liquid rubber is initially miscible with the
uncured epoxy resin, while during the curing stage it irreversibly reacts with the epoxy
through the functional groups (e.g. carboxylic or amino) and, as the molecular weight
increases, it results in the formation of a second phase [24].
Herein we report an innovative procedure aimed at introducing rubber toughening to the
polyketone/bis-maleimide system mentioned above, without interfering in the reversible
character of the thermoset polymer. The rubber, an EPM grafted with furan groups, was
chosen as a second cross-linked toughening phase as it can be easily prepared from
commercial maleated EPM and reversibly cross-linked with bis-maleimide, as recently
reported [25]. The furan-modified polyketone and EPM were blended in solution, at different
weight ratios, and cross-linked with a bis-maleimide. The corresponding morphologies were
investigated by means of electronic and optical microscopy whereas polymer toughness was
determined in terms of Izod impact strength. Thermal reversibility was tested by FT-IR and
DSC measurements while thermomechanical properties, reworkability and recyclability were
evaluated by DMTA analysis.
58
4.2 Experimental section
Reagents
The alternating aliphatic polyketone (PK30, MW 2687 Da) presents a total olefin content of
30% of ethylene and 70% of propylene [26, 27]. Maleated ethylene/propylene rubber (EPM-
MA, 49 wt.% ethylene, 49 wt.% propylene, 2.1 wt.% maleic anhydride, Mw = 50 kg/mol, PDI
= 2.0) was kindly provided by LANXESS elastomers and dried in a vacuum oven at 175 °C
for one hour to convert hydrolyzed diacids into anhydrides. Furfurylamine (Sigma-Aldrich,
≥99%) was freshly distilled before use. Tetrahydrofuran (THF, Sigma-Aldrich, >99.9%),
acetone (Sigma-Aldrich, >99.5%) and 1,1’-(methylenedi-4,1-phenylene)bis-maleimide
(Sigma-Aldrich, >95%) were purchased and used as received. Deuterated dimethylsulfoxide
(DMSO-d6, Sigma Aldrich, ≥99 atom%) was used as solvent for 1H-NMR measurements.
Furan-derivatization of PK30 via Paal-Knorr reaction
The reaction between PK30 and furfurylamine was carried out in bulk according to the Paal-
Knorr reaction (Figure 4.1).
Figure 4.1 Modification of the aliphatic polyketone in bulk via Paal-Knorr reaction.
In a typical experiment, about 60 g of PK30 were weighed into a sealed 250 ml round bottom
glass reactor equipped with a U-type anchor impeller, a reflux condenser and an oil bath for
heating. The reactor was heated to 110 °C and 35.4 g (0.365 mol) of furfurylamine, based on
the targeted carbonyl conversion of 80% (i.e. the di-carbonyl unit of polyketone modified
into pyrrole ring, Figure 4.1), were added dropwise during the first 20 minutes. The reaction
was allowed to proceed for 4 hours. The light brown product (PK-Fu) was cooled with liquid
59
nitrogen, crushed and ground into small particles. The particles were washed three times with
Milli-Q water to remove the unreacted furfurylamine, if any, filtered and freeze-dried for 24h.
The carbonyl conversion (CCO) of PK was calculated according as follows:
𝐶𝐶𝑂 =𝑦
𝑥+𝑦∙ 100 (1)
being x the moles of dicarbonyl unit and y the moles of pyrrolic unit (Figure 4.1) after
conversion. y was calculated according to the formula:
𝑦 =𝑁
𝑀𝑁 (2)
where N is the weight in grams of nitrogen in the final product, as determined by elemental
analysis and MN is the atomic mass of nitrogen. x was calculated as follows:
𝑥 =
𝐶
𝑀𝐶 − 𝑦 ∙ 𝑛𝑦
𝐶
𝑛𝑥𝐶
(3)
where C is the weight (grams) of carbon in the final product as determined by elemental
analysis, MC is the atomic mass of carbon, and are the average number of carbons in
the pyrrolic and in the dicarbonyl unit, respectively. Conversion efficiency (η) was calculated
as follows:
𝜂 =𝑦
𝑚𝑜𝑙𝐹𝑎∙ 100 (4)
with molFa being the moles of furfurylamine in the feed.
Furan-derivatization of EPM-MA
Typically 100.0 g of EPM-MA (21.42 mmol MA) were dissolved in 900 g THF by stirring
for 24 h at room temperature in a closed flask. 6.24 g of furfurylamine (3 equivalents based
on the molar content of maleic anhydride) were added to the solution and agitation was
maintained for 12 h. After reaction, the product (EPM-Fu) was purified by mixing the
polymer with 5 liters of acetone under mechanical stirring. Once the polymer precipitates as
yellowish flakes, it is collected and dried in an oven at 50 °C up to constant weight.
Conversion (C) of MA groups was calculated from FT-IR spectra according to a reported
method [25], by using the following formula:
C
yn C
xn
60
𝐶 = (1 −𝐴1856
𝐹𝑢 𝐴723𝐹𝑢⁄
𝐴1856𝑀𝐴 𝐴723
𝑀𝐴⁄) ∙ 100 (5)
where: and are the integrals of the absorption bands at 1856 cm-1
(C=O
asymmetric stretching of anhydride rings) for EPM-Fu and EPM-MA respectively; and
are the integrals of the bands at 723 cm-1
(methyl rocking), used as internal reference.
Samples preparation and cross-linking
PK-Fu and EPM-Fu (set at different weight ratios) were dissolved in a nine-fold amount of
THF, in a round bottom flask equipped with a magnetic stirrer, an oil bath and a reflux
condenser. The mixture was heated at 50 °C for 24 h. A stoichiometric amount of 1,1’-
(methylenedi-4,1-phenylene)bis-maleimide (considering a 1:1 molar ratio between maleimide
and the total amount of furan groups) was added to the mixture and the reaction (Diels-Alder
cycloaddition) was allowed to proceed for 24 h under agitation at 50 °C. After reaction, the
cross-linked gels were placed in a vacuum oven during 24 h at 50 °C to evaporate the solvent.
Reference samples, one containing PK-Fu and EPM-MA (80:20 wt. % ratio) and another
containing only PK-Fu, were prepared and cross-linked with bis-maleimide using the same
procedure. An overview of all prepared systems is reported in Table 4.1. Samples have been
coded by stating the type of EPM rubber and the percentage weight ratio rubber/polyketone.
Table 4.1 Amounts of polyketone (PK-Fu), EPM rubber (EPM-Fu or EPM-MA) and 1,1’-
(methylenedi-4,1-phenylene)bis-maleimide used to prepare the samples and resulting wt. % of rubber
in each system.
Sample Code PK-Fu
(g)
EPM
(g)
PK-Fu/EPM
(w/w)
bis-maleimide
(g)
Rubber
(wt. %)
PK-Fu* 6.00 - 100:0 4.77 (13.3 mmol) -
PK-Fu/EPM-Fu_10 5.41 0.60 90:10 4.31 (12.0 mmol) 6
PK-Fu/EPM-Fu_15 5.09 0.90 85:15 4.09 (11.4 mmol) 9
PK-Fu/EPM-Fu_20 4.80 1.20 80:20 3.86 (10.8 mmol) 12
PK-Fu/EPM-MA_20* 4.81 1.20 80:20 3.81 (10.6 mmol) 12
*Reference systems
1856FuA 1856
MAA
723FuA
723MAA
61
Characterization
Elemental analysis was performed with an EuroVector EA apparatus. 1H-NMR spectra were
recorded on a Varian Mercury Plus 400 MHz apparatus using DMSO-d6 as solvent. FT-IR
spectra were obtained using a Perkin-Elmer Spectrum 2000. KBr pellets with 1.5 wt. %
content of polymer were prepared from powders of the polyketone/rubber samples. Films (1
mm thick) were prepared from the EPM rubbers by compression molding at 180 °C for 30
minutes under a pressure of 100 bar. DSC thermograms were recorded on a TA-Instrument
DSC 2920 under N2 atmosphere. Samples were first heated from 0 °C to 180 °C and then
cooled to 0 °C. Four cycles were performed at a rate of 10 °C per minute. Dynamic
Mechanical Thermal Analysis (DMTA) was performed using a Rheometric scientific solid
analyzer (RSA II) under air environment, in dual cantilever mode at an oscillation frequency
of 1 Hz and a heating rate of 3 °C per minute. Three cycles were performed for each sample,
between room and softening temperature. DMTA specimens were 6 mm wide, 1.4 mm thick
and 54 mm long. They were prepared by molding the cross-linked samples at 150 °C for 30
minutes under a pressure of 40 bar and then annealed in an oven at 50 °C for 24 hours. Izod
impact strength was measured at room temperature, according to ASTM D4812 using
standard unnotched specimens (12.7 mm wide, 3.3 mm thick and 64 mm long) prepared by
compression molding of about 3.5 g of material in the same conditions used for DMTA
specimens. Tests were performed on a Zwick 5102 Pendulum Impact Tester equipped with a
hammer. At least 6 specimens for each sample were tested. Impact strength was calculated as
the ratio between the energy absorbed in the impact and the thickness of the specimen. The
morphological study of the systems was performed using a Philips XL30 Field-Emission
Environmental Scanning Electron Microscopy SEM-FEG on fresh-fractured surfaces coated
with gold nanoparticles (10 nm size). A simple observation on fracture toughness was
registered using an optical microscope on samples prepared with the polymerography
technique [28, 29].
4.3 Results and discussion
Synthesis of furan-functionalized polyketone
Furan-functionalized polyketone (PK-Fu) was synthesized by reacting PK30 with
furfurylamine in bulk at 110 °C for 4 h (Figure 4.1). A high quantitative carbonyl conversion
of 76% (80% targeted) and a conversion efficiency of 96% were obtained, confirming the
62
Paal-Knorr reaction as a simple and efficient route to modify polyketones, as previously
reported [11, 12, 30]. 1H-NMR spectra of PK30 and PK-Fu are shown in Figure 4.2.
Figure 4.2 1H-NMR spectra of unmodified polyketone (PK30, black curve) and furan modified
polyketone (PK-Fu, red curve).
The resonance signals at 7.5, 6.3 and 6.1 ppm were assigned to protons on the furan ring
connected to the polymer backbone, the resonance peak at 5.6 ppm was assigned to the
protons of the pyrrole ring whereas the peak at 4.9 ppm was ascribed to the CH2 group
connecting the furan and the pyrrole ring [11, 12].
Synthesis of furan-functionalized EPM rubber
A pre-thermal treatment was performed on the EPM-MA rubber in vacuum at 175 °C in order
to turn the hydrolyzed diacids moieties into maleic anhydrides [25, 31]. Subsequently, furan-
functionalized EPM rubber (EPM-Fu) was synthesized by reacting EPM-MA and
furfurylamine in a THF solution at room temperature. The resulting EPM-Fu rubber was then
studied by FT-IR transmission (Figure 4.3). The appearance of the C-O-C symmetric
stretching band at 1013 cm-1
clearly proved the presence of furan groups in the product [32].
Furthermore, the shift of the band at 1856 cm-1
to 1780 cm-1
(C=O asymmetric stretching)
63
and the one from 1780 cm-1
to 1710 cm -1
(C=O stretching) in addition to the appearance of a
new band at 1378 cm-1
(C-N symmetric stretching) demonstrated the conversion of anhydride
rings into imide derivatives [12, 31, 33].
Figure 4.3 FT-IR spectra of EPM-MA and the modified EPM-Fu: the reaction sequence is displayed
in the inset.
A deconvolution analysis was used to integrate the areas related to the peaks at 1856 cm-1
(C=O asymmetric stretching) and 723 cm-1
(methyl rocking vibration). From their ratio, the
reaction conversion was determined and was found to be >99.9%, thus demonstrating that the
chemical modification process is quantitative.
Cross-linking of polyketone/rubber blends via Diels-Alder cycloaddition
PK-Fu/EPM-Fu blends were cross-linked with 1,1’-(methylenedi-4,1-phenylene)bis-
maleimide in THF solutions at 50 °C for 24 h (see Table for experimental conditions). A
scheme of one of the possible Diels-Alder (DA) adducts is reported in Figure 4.4. The FT-IR
spectra of the sample PK-Fu/EPM-Fu_20 were recorded before and after the reaction with the
bis-maleimide (Figure 4.5).
64
Figure 4.4 Cross-linking and de-cross-linking of furan-modified polyketone PK-Fu / EPM-Fu rubber
blends via Diels Alder (DA) and retro-Diels-Alder (r-DA) sequence with 1,1’-(methylenedi-4,1-
phenylene)bis-maleimide.
Figure 4.5 clearly displays the appearance of the band at 1185 cm-1
(C-N-C stretching in
succinimide ring of DA adduct) and the disappearing of the bands at 740 cm-1
(out-of-plane
proton bending) and 1011 cm-1
(Vs C-O-C) of unreacted furans. Notably, this result
substantially proves the formation of Diels-Alder adducts [11, 12, 34, 35] while the band at
1378 cm-1
(C-N stretching in maleimide rings) demonstrates the presence of the bis-
maleimide.
Figure 4.5 FT-IR spectra of PK-Fu/EPM-Fu (80/20 wt. % ratio) blend before and after cross-linking
with 1,1’-(methylenedi-4,1-phenylene)bis-maleimide.
65
All cross-linked samples were characterized by Differential Scanning Calorimetry (DSC) in
order to determine the thermal history, reversibility and the exo / endo-thermal process
related to the DA and r-DA sequence, respectively. All thermograms displayed a broad
endothermic transition in the range of temperature 120-150 °C for each consecutive thermal
cycle (Figure 4.6). The similarity in each consecutive thermal cycle demonstrates the
reversible character of the cross-linked PK-Fu, even when blended with the rubber.
Figure 4.6 DSC thermal cycles of cross-linked PK-Fu/EPM-Fu_20. Only one sample was reported for
clarity.
The endothermic transition corresponds to the r-DA process. Therefore, the peak of the
curves corresponds to the temperature at which the majority of the DA adducts are broken
(TrDA peak). The area under the curve associated to this peak, is therefore related to the energy
absorbed during the cleavage of the DA adducts [7, 12]. Remarkably, it can be noticed that
the energy absorbed during the r-DA process (Table 4.2) decreases as the rubber content
increases in the systems. This trend can be explained considering that EPM-Fu has a lower
moles number of furan groups per gram than the PK-Fu, thus less DA adducts are present in
the blends compared to the neat cross-linked polyketone. Indeed, the lowest value in the
0 20 40 60 80 100 120 140 160 180
-6
-5
-4
-3
-2
-1
0
1
2
Cycles
Hea
t fl
ow
(W
/g)
end
o u
p
Temperature (°C)
1
2
3
0 20 40 60 80 100 120 140 160 180
-6
-5
-4
-3
-2
-1
0
1
2
Cycles
Hea
t fl
ow
(W
/g)
end
o u
p
Temperature (°C)
1
2
3
66
integral was obtained from the sample PK-Fu/EPM-MA_20, which contains the maleated
EPM rubber. All systems containing the rubber displayed practically the same TrDA peak and
about eight degrees higher than the reference system PK-Fu. This can be possibly explained
by the absorption of energy from the rubber during the process of r-DA and disentanglement
[36, 37].
Table 4.2 Calculated endothermic integral values (ArearDA) and r-DA endothermic peak TrDA in the 1st
thermal cycle of PK-Fu, PK-Fu/EPM-Fu and PK-Fu/EPM-MA cross-linked with bis-maleimide.
Sample Area rDA
(J/g)
TrDA peak
(°C)
PK-Fu* 19.74 141.8
PK-Fu/EPM-Fu_10 17.33 149.6
PK-Fu/EPM-Fu_15 16.68 149.7
PK-Fu/EPM-Fu_20 15.38 150.4
PK-Fu/EPM-MA_20* 13.31 150.4
*Reference samples
Homogeneous test-specimens for Dynamic-mechanical Thermal Analysis (DMTA) were
successfully obtained from all samples, using hot compression molding. The variations in
storage modulus (E’) and softening point (tan δ) as a function of the temperature and different
amounts of rubber are shown in Figure 4.7.
For all samples the storage modulus (E’) exhibited a plateau followed by a sharp drop starting
around 110 °C, implying a rubbery transition, but still cross-linked, due to a progressive
softening of the entire system that probably loses some of the DA adducts during the process
[8]. E’ values in the plateau region were in the same order of magnitude for all the systems,
but a lower value (1.5 GPa) was observed for the sample with the highest content of rubber
(PK-Fu/EPM-Fu_20), thus suggesting a less stiff material. The softening temperature, as
determined from the peak of tan δ curves, was found within a narrow range (140-146 °C) for
PK-Fu, PK-Fu/EPM-Fu_10, PK-Fu/EPM-Fu_15 and PK-Fu/EPM-MA_20, while
significantly higher (155 °C) for PK-Fu/EPM-Fu_20.
67
Figure 4.7 Storage modulus E’ (A) and softening point tan δ (B) in DMTA during 1st heating stage
for cross-linked PK-Fu, PK-Fu/EPM-Fu_10, PK-Fu/EPM-Fu_15 and PK-Fu/EPM-Fu_20.
After reaching the softening temperature, each sample was cooled down to room temperature
during the first 10 minutes inside the DMTA chamber and tested again up to 3 heating cycles
(Figure 4.8).
Storage and loss moduli were quantitatively recovered in each cycle, providing further
evidence of the reworkability of the systems. Notably, the softening temperature slightly
shifted to higher values (+5 °C) for samples containing the rubber. This shift can be attributed
to a change from the kinetically endo to the thermodynamically more stable exo conformation
of DA adducts after heating cycles [38].
30 40 50 60 70 80 90 100 110 120 130 140 150 16010
7
108
109
PK-Fu
PK-Fu/EPM-Fu_10
PK-Fu/EPM-Fu_15
PK-Fu/EPM-Fu_20
E' (
Pa)
Temperature (°C)
30 40 50 60 70 80 90 100 110 120 130 140 150 1600,00
0,05
0,10
0,15
0,20
0,25
PK-Fu
PK-Fu/EPM-Fu_10
PK-Fu/EPM-Fu_15
PK-Fu/EPM-Fu_20
Tan
()
Temperature (°C)
68
Figure 4.8 Dynamic Mechanical Thermal Analysis of cross-linked samples PK-Fu (A), PK-
Fu/EPMFu_10 (B) and PK-Fu/EPM-Fu_20 (C) in three consecutive cycles.
30 40 50 60 70 80 90 100 110 120 130 140 150 160 170
107
108
109
1
2
3
Temperature (oC)
E'E
'' (P
a)
0,00
0,05
0,10
0,15
0,20
0,25
Tan
()
tan(
E''
E'
30 40 50 60 70 80 90 100 110 120 130 140 150 160 170
107
108
109
Temperature (°C)
E'E
'' (P
a)
0,00
0,05
0,10
0,15
0,20
0,25
tan(
E''
E'
Tan
()
1
2
3
30 40 50 60 70 80 90 100 110 120 130 140 150 160 170
107
108
109
Temperature (oC)
E'E
'' (P
a)
0,00
0,05
0,10
0,15
0,20
0,25
tan(
E''
E' 1
2
3
Tan
()
69
Figure 4.8 also shows that the presence of increasing amounts of rubber amplifies the system
ability to recover loss modulus between cycles. According to this, the presence of rubber
increases the system dimensional stability so that modulus remains practically as the original
in continuous DMTA measurements when the concentration of rubber reached 12 % (PK-
Fu/EPM-Fu_20). After thermo-mechanical tests, all samples were re-grinded and reshaped by
compression molding for further DMTA tests, in order to further check their recyclability
(Figure 4.9).
Figure 4.9 DMTA of cross-linked samples PK-Fu (A) and PK-Fu/EPM-Fu_20 (B): comparison
between virgin (freshly molded, in black) and recycled (grinded and remolded, in green) material. The
recycled samples were reshaped at 150 °C and 40 bar for 30 minutes.
40 60 80 100 120 140 160 180 200
106
107
108
109
virgin
recycled
Temperature (°C)
E'E
'' (P
a)
E'
E''
tan(
0,00
0,05
0,10
0,15
0,20
0,25
Tan
()
40 60 80 100 120 140 160 180 200
106
107
108
109
virgin
recycled
Temperature (oC)
E'E
'' (P
a)
0,00
0,05
0,10
0,15
0,20
0,25
Tan
()E''
E'
tan(
70
It is worth noticing that both recycled samples displayed quantitative recovery of modulus
and softening points. However a slightly increase in softening point was observed for the neat
PK-Fu thermoset sample, possibly due to a more effective endo to exo conformational change
of the DA adducts (Figure 4.9). The relatively high softening point and the fully quantitative
recovery of modulus and tan δ during thermo-mechanical cycles and recycling (grained
samples), denotes that fully thermally reversible systems are possible to be made by the
combination of furan functionalized polyketone with furan functionalized EPM rubber.
Izod Impact test was performed on unnotched specimens that broke completely during testing
(Figure 4.10).
Figure 4.10 Izod impact strength of all polymer samples
All toughened samples display a monotonous increase in the impact strength with increased
rubber content. For the sample PK-Fu/EPM-Fu_20 (12 wt. % of furan-modified rubber) a
twofold increase was observed compared to the neat cross-linked polyketone. Moreover, a
slightly higher value of impact strength for PK-Fu/EPM-Fu_20 compared to the reference
PK-Fu/EPM-MA (12 wt. % of maleated rubber) was observed. This small difference can
probably indicate that both EPM rubbers provide similar toughening effect on the cross-
linked polyketone. However, the DA chemical interaction between the matrix and the furan-
PK-Fu PK-Fu/
EPM-Fu_10
PK-Fu/
EPM-Fu_15
PK-Fu/
EPM-Fu_20
PK-Fu/
EPM-MA_20
0
5
10
15
20
25
30
35
Izo
d I
mp
act
stre
ng
th (
J/m
)
71
functionalized rubber may be also taken into consideration. In order to clarify this point, a
morphological study was performed by means of scanning electron microscopy (SEM) on
fresh fractured surfaces in both toughened systems (Figure 4.11). The holes observed in the
micrographs are cavitated spaces of detached rubber particles after the samples are broken. It
is clearly noticed that EPM-MA and EPM-Fu rubber particles are embedded heterogeneously
in size and morphology inside the matrix (Figure 4.11A and 4.11B). However, EPM-MA
rubber particles leave smoother surfaces inside and around holes when removed, therefore
suggesting poor compatibility and no chemical bonds between the two phases (Figure 4.11C).
Differently, EPM-Fu rubber particles leave rough surfaces inside and around holes suggesting
a well intrinsic adhesion between the phases. This fact might be explained by the interfacial
chemical bond via DA active groups (Figure 4.11D).
Figure 4.11 SEM micrographs of freshly broken surfaces of PK-Fu/EPM-MA_20 (A and C) and PK-
Fu/EPM-Fu_20 (B and D) showing cavitated particles (indicated by white arrows) with different size
and morphology.
72
Figure 4.12 displays the toughening effect of EPM-MA and EPM-Fu rubber particles in the
thermoset matrix after samples break. In Figure 4.12B, a crack that apparently propagates
from right to left (black arrow) is deflected and bridged by EPM-MA rubber particles (red
arrow). Moreover, some rubber particles are pulled-out or separated from the matrix as the
crack propagates through the matrix. This can be explained by the weaker interaction
between both phases as observed in Figure 4.11C and suggested by impact test results. On the
other hand, Figures 4.12C and 4.12D show that crack propagation is stopped by EPM-Fu
rubber particles. Deflection is also observed (black arrow, Figure 4.12D) but crack separation
seems to be absent. Overall, this demonstrates the effective intrinsic adhesion between the
phases possibly due to the presence of DA inter-linkers between them.
Figure 4.12 Optical micrographs of PK-Fu (A) PK-Fu/EPM-MA_20 (B) and PK-Fu/EPM-Fu_20 (C,
D) at different magnifications displaying the toughening effect of rubber particles. Letter R indicates
rubber particles.
A
C
Artefact
B
DR
R
R
R
R
R
73
4.4 Conclusions
We have demonstrated the facile synthesis of a reversible thermoset material based on the
chemical modification of aliphatic polyketones with furfurylamine by the Paal-Knorr
reaction. The cross-linking and the rubber toughening with functionalized furan-EPM rubber
occur via Diels-Alder chemistry. Moreover, all the reaction involving polymers
functionalization occur without adding any catalyst and in a conventional apparatus thus
suggesting scaling up feasibility. Additionally, this approach successfully demonstrated
thermo-mechanical, impact strength and recycling improvements on the neat thermoset. The
combination of DSC and DMTA experiments effectively indicates that furan functionalized
polyketones in combination with furan-functionalized rubber are capable to be repeatedly
cross-linked and de-cross-linked with bis-maleimide by only using heat as external stimulus.
Impact strength and morphological studies clearly demonstrated that the introduction of
covalently bonded EPM rubber via Diels-Alder chemistry effectively improved the final
impact strength of the toughened thermoset systems.
4.5 References
1. Ratna D. Handbook of thermoset resins: ISmithers, 2009.
2. Guo Q. Thermosets structure, properties and applications. Cambridge, UK; Philadelphia,
PA: Woodhead Publishing, 2012.
3. Pickering SJ. Recycling technologies for thermoset composite materials—current status.
Composites Part A: Applied Science and Manufacturing 2006;37:1206-1215.
4. Thomas R, Vijayan P, Thomas S. 4. Recycling of thermosetting polymers: Their blends
and composites. Recent Developments in Polymer Recycling 2011;121:153.
5. Yang Y, Boom R, Irion B, van Heerden D, Kuiper P, de Wit H. Recycling of composite
materials. Chemical Engineering and Processing: Process Intensification 2012;51:53-68.
6. Kloxin CJ, Bowman CN. Covalent adaptable networks: smart, reconfigurable and
responsive network systems. Chem. Soc. Rev. 2013;42:7161-7173.
7. Scheltjens G, Brancart J, De Graeve I, Van Mele B, Terryn H, Van Assche G. Self-healing
property characterization of reversible thermoset coatings. Journal of Thermal Analysis and
Calorimetry 2011;105:805-809.
8. Gandini A. The furan/maleimide Diels–Alder reaction: A versatile click–unclick tool in
macromolecular synthesis. Progress in Polymer Science 2013;38:1-29.
9. Watanabe M, Yoshie N. Synthesis and properties of readily recyclable polymers from
bisfuranic terminated poly(ethylene adipate) and multi-maleimide linkers. Polymer
2006;47:4946-4952.
10. Gheneim R, Perez-Berumen C, Gandini A. Diels-Alder reactions with novel polymeric
dienes and dienophiles: Synthesis of reversibly cross-linked elastomers. Macromolecules
2002;35:7246-7253.
11. Zhang Y, Broekhuis AA, Picchioni F. Thermally Self-Healing Polymeric Materials: The
Next Step to Recycling Thermoset Polymers?. Macromolecules 2009;42:1906-1912.
74
12. Toncelli C, De Reus DC, Picchioni F, Broekhuis AA. Properties of Reversible Diels-
Alder Furan/Maleimide Polymer Networks as Function of Crosslink Density.
Macromolecular Chemistry and Physics 2012;213:157-165.
13. Bagheri R, Marouf BT, Pearson RA. Rubber-Toughened Epoxies: A Critical Review.
Polymer Reviews 2009;49:201-225.
14. Nakamura Y, Yamaguchi M, Okubo M, Matsumoto T. Effect of particle-size on the
fracture-toughness of epoxy-resin filled with spherical silica. Polymer 1992;33(16):3415-
3426.
15. Lee J, Yee AF. Inorganic particle toughening II: toughening mechanisms of glass bead
filled epoxies. Polymer 2001;42(2):589-597.
16. Pearson RA, Yee AF. Toughening mechanisms in thermoplastic-modified epoxies. 1.
Modification using poly(phenylene oxide). Polymer 1993;34(17):3658-3670.
17. Rusli A, Cook WD, Schiller TL. Blends of epoxy resins and polyphenylene oxide as
processing aids and toughening agents 2: Curing kinetics, rheology, structure and properties.
Polym Int 2014;63(8):1414-1426.
18. Giannakopoulos G, Masania K, Taylor AC. Toughening of epoxy using core-shell
particles. Journal of Materials Science 2011;46(2):327-338.
19. Chen J, Kinloch AJ, Sprenger S, Taylor AC. The mechanical properties and toughening
mechanisms of an epoxy polymer modified with polysiloxane-based core-shell particles.
Polymer 2013;54(16):4276-4289.
20. Kinloch AJ, Lee SH, Taylor AC. Improving the fracture toughness and the cyclic-fatigue
resistance of epoxy-polymer blends. Polymer 2014;55(24):6325-6334.
21. Ozturk A, Kaynak C, Tincer T. Effects of liquid rubber modification on the behaviour of
epoxy resin. European Polymer Journal 2001;37(12):2353-2363.
22. Chikhi N, Fellahi S, Bakar M. Modification of epoxy resin using reactive liquid (ATBN)
rubber. European Polymer Journal 2002;38(2):251-264.
23. Tripathi G, Srivastava D. Effect of carboxyl-terminated poly (butadiene-co-acrylonitrile)
(CTBN) concentration on thermal and mechanical properties of binary blends of diglycidyl
ether of bisphenol-A (DGEBA) epoxy resin. Materials Science and Engineering a-Structural
Materials Properties Microstructure and Processing 2007;443(1-2):262-269.
24. Ratna D, Banthia AK, Deb PC. Acrylate-based liquid rubber as impact modifier for epoxy
resin. Journal of Applied Polymer Science 2001;80(10):1792-1801.
25. Polgar LM, van Duin M, Broekhuis AA, Picchioni F. The use of Diels-Alder chemistry
for thermo-reversible cross-linking of rubbers: the next step towards recycling of rubber
products?. DOI: 10.1021/acs.macromol.5b01422.
26. Drent E, Keijsper JJ. Liquid polymers 1993.
27. Mul WP, Dirkzwager H, Broekhuis AA, Heeres HJ, van der Linden AJ, Orpen AG.
Highly active, recyclable catalyst for the manufacture of viscous, low molecular weight, CO-
ethene-propene-based polyketone, base component for a new class of resins. Inorganica
Chimica Acta 2002;327:147-159.
28. Bartosiewicz L, Kelly C. Microscopy techniques for polymer characterization. Marcel
Dekker, Inc., Encyclopedia of Engineering Materials.Part A: Polymer Science and
Technology. 1988;1:537-604.
29. Battaerd H. The significance of incompatible polymer systems 1975;49(1):149-157.
30. Araya-Hermosilla R, Broekhuis AA, Picchioni F. Reversible polymer networks
containing covalent and hydrogen bonding interactions. European Polymer Journal
2014;50:127-134.
75
31. van der Mee MAJ. Thermoreversible cross-linking of elastomers : a comparative study
between ionic interactions, hydrogen bonding and covalent cross-links. Eindhoven:
Technische Universiteit Eindhoven, 2007.
32. Liu X, Du P, Liu L, Zheng Z, Wang X, Joncheray T, Zhang Y. Kinetic study of Diels-
Alder reaction involving in maleimide-furan compounds and linear polyurethane. Polymer
Bulletin 2013;70:2319-2335.
33. Vermeesch I, Groeninckx G. Chemical Modification of Poly(styrene-Co-Maleic
Anhydride) with Primary N-Alkylamines by Reactive Extrusion. Journal of Applied Polymer
Science 1994;53:1365-1373.
34. Gaina C, Ursache O, Gaina V, Varganici CD. Thermally reversible cross-linked
poly(ether-urethane)s. Express Polym Lett 2013;7(7):636-650.
35. Ursache O, Gaina C, Gaina V, Tudorachi N, Bargan A, Varganici C, Rosu D. Studies on
Diels-Alder thermoresponsive networks based on ether-urethane bismaleimide functionalized
poly(vinyl alcohol). J Therm Anal Calorim 2014;118(3):1471-1481.
36. Twardowski T, Kramer O. Elastic Contributions from Chain Entangling and Chemical
Cross-Links in Elastomer Networks in the Small-Strain Limit. Macromolecules
1991;24(21):5769-5771.
37. Kluppel M. Finite Chain Extensibility and Topological Constraints in Swollen Networks.
Macromolecules 1994;27(24):7179-7184.
38. Xu Z, Zhao Y, Wang X, Lin T. A thermally healable polyhedral oligomeric
silsesquioxane (POSS) nanocomposite based on Diels-Alder chemistry. Chemical
Communications 2013;49:6755-6757.
76
5 Exfoliation and stabilization of MWCNTs in a thermoplastic pyrrole-containing
matrix assisted by hydrogen bonds
Abstract. This work focuses on the design of an engineered thermoplastic polymer
containing pyrrole units in the main chain and hydroxyl pendant groups (A-PPy-OH), which
help in achieving nanocomposites containing well-distributed, exfoliated and undamaged
multi-walled carbon nanotubes (MWCNTs). The thermal annealing at 100 °C of the pristine
nanocomposite promotes the redistribution of the nanotubes in terms of a percolative
network, thus converting the insulating material in a conducting soft matrix (60 μ Ω.m). This
network remains unaltered after cooling to r.t. and to successive heating cycles up to 100 °C
thanks to the effective stabilization of MWCNTs provided by the functional polymer matrix.
Notably, the resistivity–temperature profile is very reproducible and with a negative
temperature coefficient of -0.002 K-1
, which suggests the potential application of the
composite as a temperature sensor.
Keywords: thermoplastic pyrrole-containing polymer, MWCNTs exfoliation, MWCNTs
thermodynamic stabilization, hydrogen bonding, temperature sensor.
77
5.1 Introduction
Carbon nanotubes (CNTs) are one of the most popular fillers currently used in polymer
nanocomposites (PNCs) [1]. CNTs are cylindrical forms of graphitic sheets displaying a
single wall (SWCNT) or multi walls (MWCNT) with open or closed ends [2]. CNTs are
inherently multifunctional so they can work as structural support [3], conductive [4, 5] and
sensing platform in PNCs [6]. Conductive nanocomposites have been currently used in
several commercial products [7]. Among them, PNCs with temperature sensing properties
have become very attractive products in the open market due to their possibility of nanoscale
tailoring and very low cost of production. Typically, these kinds of materials have shown a
resistivity strictly dependent on temperature [8], thus opening successful applications in the
field of miniaturized and potentially low cost plastic sensors [6, 9-12]. In some cases, the
resistance variation proceeds by the dynamic interconnection/disconnection of the CNT
network in the matrix [13]. However, temperature sensing properties of CNTs/polymer
composites are demonstrated to be more reproducible when the resistive response is governed
by the semi-conducting features of exfoliated and stabilized CNT networks under thermal
solicitations [11].
Despite all advantages of electronic temperature-sensing PNCs over conventional
thermometers, these kinds of materials are not exempt of problems. A common drawback in
the design of resistive sensors based on CNTs/polymer composites is the strong tendency of
CNTs to aggregate in bundles during composite processing due to the strong van der Waals
interactions between their graphitic surfaces, which make their large-scale utilization
problematic [7, 14, 15].
Several strategies to improve CNT dispersion in polymeric matrices have been reported in the
open literature (e.g. CNT/in-situ polymerization composites, high-shear melts processing,
injection molding etc. [16]). Among them, particularly attractive is the functionalization of
the CNTs surfaces by the covalent attachment of functional groups [3, 16-18], but disruption
of their sp2 conducting network even occurs [19, 20]. A non-disruptive strategy used to
disperse CNTs by means of conductive polypyrroles has been reported for applications in
electronics [21, 22]. In these particular systems, the pyrrole groups get in contact with the sp2
network of the CNTs surface via supramolecular π-π interactions, which promote the polymer
wrapping around the filler. As a result, multilayers of the polymer form bridges that separate
78
CNTs from each other yielding effective percolation pathways. However, despite the good
conductive properties, polypyrroles/MWCNTs composites are highly brittle so that they must
be doped with counter ions or coated on flexible polymer substrates to improve their
mechanical performance (e.g. fracture toughness)[23, 24].
Herein we report on the design of a thermoplastic OH-functionalized alternating aliphatic
polypyrrole (A-PPy-OH) matrix, which is capable to exfoliate and stabilize MWCNTs
without the need of surface modification of the filler. The thermoplastic polymer is designed
by the chemical modification of alternating aliphatic polyketone PK via Paal-Knorr reaction
with amine compounds. The production of the polymer [25] and its chemical modification
[26] occurs in high yield, low cost and fast kinetic using relatively mild conditions in bulk
and with water as the only by-product. Specifically, the reaction between the polymer and an
OH-amine compound turns the alternating carbonyl backbone of PK into pyrrole units
bearing hydroxyl moieties (Figure 5.1). The combination of the polymer with MWCNTs
produces a malleable and conductive rubber-like nanocomposite displaying electronic
temperature-sensing properties. On one hand, the pyrrolic backbone exfoliates and stabilizes
bundles of MWCNTs via supramolecular π-π interactions with the graphitic surface of the
filler during thermal annealing. On the other hand, the OH-functional groups pending from
the pyrrole units assist the polymer during the thermal stress by hydrogen bonding
interactions to keep its dimensional stability and mechanical features. In order to figure out
the role of the OH-motifs, a reference polymer displaying the same pyrrolic backbone of A-
PPy-OH was prepared by the chemical modification of PK with n-butylamine, where the
hydroxyl group is replaced by a CH3 group (A-PPy-CH3). The CNTs exfoliation and their
effective dispersion within polymer matrices were investigated by in-situ resistance
measurements and charge-contrast SEM imaging, whereas composites resilience under
thermal stress was investigated by DSC analysis. The electrical resistance of composites was
eventually evaluated under thermal cycles between r.t. and 100 °C, to explore their potential
for the development of sensitive, stable, and reproducible temperature sensors.
79
Figure 5.1 Schematic representation of PK chemically modified with 3-amino-1-propanol via the
Paal-Knorr reaction (A-PPy-OH), and the resulting composite after mixing with MWCNTs.
5.2 Experimental section
Reagents
The alternating aliphatic polyketone (PK30, MW 2687 Da, Mul et al. 2002 [27]) presents a
total olefin content of 30% of ethylene and 70% of propylene. 3-amino-1-propanol (OH)
(Acros), butylamine (CH3) (Sigma Aldrich, 99%), 2,5-hexanedione (Sigma Aldrich 98%),
multi-walled carbon nanotubes MWCNTs (O.D. 6-9 nm, L. 5 µm, Sigma-Aldrich 95%
carbon), DMSO-D6 (Laboratory-Scan, 99.5%), 1-methyl-2-pyrrolidinone (Sigma-Aldrich,
99.5%) were purchased and used as received.
Model reaction
A model reaction between stoichiometric amounts of 2,5-hexanedione (8,7 mmol) and 3-
amino-1-propanol was carried out in order to identify the presence of any side product after
the Paal-Knorr reaction [26]. The reaction was performed in bulk, in a 100 mL round-bottom
flask equipped with a reflux condenser and a magnetic stirrer. The reaction mixture was
heated up to 100 °C during 4 h.
Functionalization of polyketone with alcohol pendant groups
The solvent-free Paal-Knorr reaction [26] between PK and 3-amino-1-propanol was carried
out using different molar ratios between the 1,4-di-carbonyl groups of polyketone and 3-
amino-1-propanol (Table 5.1). A reference polymer that displays the same backbone
Flexible and temperature-
responsive nanocomposite
80
structure, but bearing a CH3 instead of an OH group was also prepared using butylamine
instead of 3-amino-1-propanol in the Paal-Knorr reaction, with the aim of evaluating the
effect of hydrogen bonds on the thermal resilience of the composite. These chemical
modifications of the polyketone were carried out in a 250 mL round-bottom glass reactor
equipped with a reflux condenser, a U-type anchor impeller and an oil bath for heating. First,
60 g of PK (0.455 moles of di-carbonyl unit) were preheated to a liquid state at 100 °C. Then,
3-amino-1-propanol or butylamine was added dropwise to the reactor during 20 min. Next,
the stirring speed was set to 600 rpm and the reaction was carried out for 4 h. Initially, the
reaction mixture was colourless, but gradually turned to brown due to pyrrole formation on
the polymer backbone [26, 28, 29]. The resulting mixture was diluted with chloroform and
washed 3 times with a 0.2 M NaCl Milli-Q water solution, to remove the unreacted 3-amino-
1-propanol, if any. Thereafter, the organic phase was evaporated under vacuum at 50 °C for
24 h. Light-brown powders were obtained as final products. In order to avoid hydration, the
samples were sealed in brown glass vials and stored at 6 oC for further characterization. The
corresponding samples are coded as A-PPy-OHai or A-PPy-CH3ai with ai being the mol
percentage of amine with respect to the carbonyl groups in the feed. The percentage of
conversion of carbonyls (Cco) into pyrrole groups can be calculated as follows:
𝐶𝑐𝑜 =𝑦
𝑦+𝑥 ∙ 100% (1)
where x and y represent the moles of di-ketone and pyrrolic units after conversion,
respectively (Figure 5.1). y can be calculated as follows:
𝑦 =𝑤𝑡(𝑁)
𝐴𝑚 (𝑁) (2)
where wt(N) represents the grams of nitrogen in the final product according to elemental
analysis, and Am(N) is the atomic mass of nitrogen. x can be calculated as follows:
𝑥 =𝑔𝑝𝑟𝑜𝑑−𝑦∙𝑀𝑤
𝑦
𝑀𝑤 𝑝𝑘 (3)
where gprod represents the grams of product after conversion, Mwy the molecular weight of the
pyrrolic unit and Mwpk
the molecular weight of a 1,4 di-ketone unit (131,6 g/mol). The
conversion efficiency η is defined as the ratio between the carbonyl conversion Cco and the
targeted one according to the amount of polymer and amine compounds provided in the feed
(Ccofeed
):
81
𝜂 =𝐶𝑐𝑜
𝐶𝑐𝑜 𝑓𝑒𝑒𝑑 ∙ 100% (4)
the Cco feed
is calculated as follows:
𝐶𝑐𝑜 𝑓𝑒𝑒𝑑
=𝑀𝑜𝑙𝑎𝑚𝑖𝑛𝑒
𝑀𝑜𝑙𝑑−𝑐𝑜 ∙ 100% (5)
with Molamine representing the moles of amine compounds and Mold-co the moles of di-
carbonyl units in the feed.
Table 5.1 Chemical modification of PK with different amounts of 3-amino-1-propanol (OH) and
butylamine (CH3). Samples are coded by stating the target percentage of conversion of carbonyls into
pyrrole units considering a maximal 80% (ai) of modification [29].
Sample Molamine (moles) CCO a
(%) η b
(%)
A-PPy-OH20 0.091 15 75
A-PPy-OH40 0.182 36 90
A-PPy-OH60 0.274 57 95
A-PPy-OH80 0.365 79 99
A-PPy-CH380 0.365 78 98
a CCO is the % of carbonyl conversion (determined by
elemental analysis). b
η is the conversion efficiency of carbonyl groups. In all
samples 0.455 moles of di-carbonyl unit (60 g. of PK) were
used.
A-PPy-OH40 / MWCNTs composite
A-PPy-OH40 and MWCNTs were mixed in N-methylpyrrolidone, a solvent reported as an
effective dispersant for MWCNTs [17], using fixed amounts of polymer and different
amounts of MWCNT expressed as wt.%. For this step, only one of the OH-functionalized
polymers was selected aiming at the design of a flexible temperature-responsive
nanocomposite [30] (Table 5.2).
82
Table 5.2 Experimental conditions of A-PPy-OH40 mixed with different wt.% of MWCNTs.
Polymer Polymer (g) MWCNT (wt.%)
A-PPy-OH40 4 2
A-PPy-OH40 4 4
A-PPy-OH40 4 6
A-PPy-OH40 4 8
A-PPy-OH40 4 10
In detail, 4 g of polymer were completely dissolved in N-methylpyrrolidone (10 vol.%). The
required wt.% of MWCNTs was mixed with the same solvent and sonicated in a bath for 30
min and then poured to the polymer solution in a round bottomed flask at 50 °C for 24 h
under stirring. Then, the mixture was rotary evaporated and finally transferred into a vacuum
oven (80 °C for 72 h) to ensure the complete removal of the solvent.
Characterization
Elemental analysis was performed with an Euro EA elemental analyzer. It was used to
establish the percentage of nitrogen in the modified polymers. 1H-NMR spectra were
recorded on a Varian Mercury Plus 500 MHz apparatus using DMSO-d6 as solvent. ATR-
FT-IR spectra were recorded using a Thermo Nicolet NEXUS 670 FT-IR. Differential
scanning calorimetry DSC thermograms were recorded on a TA-Instrument DSC 2920 under
N2 atmosphere. The samples were weighed (10-17 mg) in an aluminium pan, which was then
sealed. The samples were first heated from -20 to 180 °C and then cooled down to -20 °C.
Four cycles were performed from -20 to 180 °C, with heating and cooling rates set to 10
°C/min. GPC measurements were performed with a HP1100 Hewlett-Packard instrument.
The equipment consists of three 300 x 7.5 mm PLgel 3 µm MIXED-E columns in series and
a GBC LC 1240 IR detector. The samples were dissolved in THF to obtain a final
concentration of 1 mg/mL. THF was used as eluent at a flow rate of 1 mL/min at a
temperature of 40 °C. The calibration was done using polystyrene as standard and the data
were determined using PSS WinGPC software. Scanning electron microscope images were
83
acquired on a Philips XL30 Environmental SEM FEG instrument as previously reported [31,
32].
In-situ resistance measurements were performed during thermal annealing (at 100 °C) on
samples (7.28 mm long, 5.85 mm wide, 1.25 mm thick) constrained between copper plates,
connected to a multimeter and placed inside a chamber provided with a heater and a
temperature controller ( 0.1 °C). Dynamic mechanical thermal analysis (DMTA) was
conducted on a Rheometrics scientific solid analyzer (RSA II) under an air environment
using the dual cantilever mode at an oscillation frequency of 1 Hz and a heating rate of 3
°C/min between 0 and 35 °C. The samples for DMTA analysis were prepared by
compression-molding of 500 mg of the composite into rectangular bars (6 mm wide, 1 mm
thick, 54 mm long) at 150 °C for 30 min under a pressure of 40 bar to ensure full
homogeneity.
5.3 Results and discussion
Model component preparation
Model compounds are useful to characterize the structure of complex polymer systems via
effective and rapid investigation techniques such as 1H-NMR spectroscopy. Here, a model
compound was prepared by reaction of 2,5-hexanedione (i.e. representative for the di-
carbonyl moieties along the PK backbone) with 3-amino-1-propanol. This Paal-Knorr
reaction resulted in the formation of a pyrrole unit, bearing a hydroxy terminated alkyl chain
(Figure 5.2).
Figure 5.2 1H-NMR spectrum of model compound prepared by reaction of 2,5-hexanedione with 3-
amino-1-propanol. The signal of protons associated to the pyrrole ring is at 5.6 ppm, CH2 units give
signals at 1.6, 3.4 and 3.8 ppm and the -OH group gives a broad signal at 4.6 ppm.
6 5 4 3 2 1
ppm
84
Paal-Knorr chemical modification of PK
The reactions between PKs and 3-amino-1-propanol (Figure 5.1) were carried out according
to the different molar ratios as reported in Table 5.1. Notably, these functionalized PKs
display a total carbonyl conversion (CCO, measured by elemental analysis) close to the target
conversion (relative efficiency, η ≥ 90%, see Table 5.1), with a slightly lower value for the A-
PPy-OH20 sample only. This result validates the robustness and versatility of the Paal-Knorr
reaction of PK. The formation of the desired modified PKs containing pyrrole units with –OH
motifs was confirmed by 1H-NMR spectroscopy (A-PPy-OH40 sample, Figure 5.3).
Figure 5.3 1H-NMR spectrum of A-PPy-OH40 in DMSO-d6.
The 1H-NMR spectrum shows analogous signals to those found in the model compound
(Figure 5.2), i.e. the pyrrole units at 5.8 ppm, the CH2 in the pendant group attached to the
pyrrole unit at 3.5 and 3.8 ppm, and the hydroxyl group at 4.65 ppm.
An increase in the ratio between the amine compound and the polyketone in the Paal-Knorr
reaction leads to the expected higher degree of conversion of the carbonyl groups of
polyketone backbone into pyrrole units, as shown by the increasing Cco values in the A-PPy-
OHai series from A-PPy-OH20 to A-PPy-OH80 (Table 5.1). The increase in the
concentration of pyrrole units along the polymer backbone promotes polymer rigidity, as
demonstrated by the enhancement in the Tg values of the functionalized polymers (Figure
5.4).
6 5 4 3 2 1
ppm
85
Figure 5.4 Tg of the polymers of the A-PPy-OHai series (in black) and of A-PPy-CH380 (in red),
plotted as a function of the degree of carbonyl conversion (Cco). The Tg values were measured by DSC
in three temperature cycles from -20 to 180 oC.
The linear increase of the Tg with Cco is not only ascribed to the presence of pyrrole units in
the backbone, but mainly to effective interactions between the -OH groups of the polymer
chains. This hypothesis is suggested by the much higher Tg value of A-PPy-OH80 (58 °C)
compared to that of A-PPy-CH380 (14 °C), which is characterized by the same conversion
degree but contains only a paraffinic moiety as pendant group. The role of -OH groups in
promoting secondary interactions among macromolecules was further investigated by FT-IR
spectroscopy (Figure 5.5).
Figure 5.5 FT-IR spectra of the A-PPy-OHai series and A-PPy-CH3ai. Numbers indicate final % of
carbonyl conversion after Paal-Knorr reaction.
20 40 60 80
0
10
20
30
40
50
60 R2 0.99
Tg (
oC
)
Cco (%)
3800 3700 3600 3500 3400 3300 3200 3100 3000
Abso
rban
ce
Wavenumber cm-1
86
The characteristic band attributed to the -OH groups stretching modes (3600-3100 cm-1)
becomes progressively wider and shifts to lower energies as the degree of functionalization of
the A-PPy-OHai polymers increases. This is likely due to the higher number of hydrogen
donors (OH) in the polymer chains. According to the literature, hydrogen donors undergo
highly directional interactions with their hydrogen acceptors [33, 34, 35]. In our case and
according to Figure 5.5, the intensity of the peak varies according to the hydrogen bonding
density, which depends on the degree of polymer modification. However, if a shift of the
peak is observed, it might be attributed to the balance between different competing
intermolecular hydrogen bonds [33]. So that, the red-shift of about 100 cm-1
observed in
Figure 5.5 suggests that one single directional interaction of the -OH group can be excluded.
This fact can be explained by the mobility of hydrogen bonding so that its interaction with
different groups on the surface of the polymer could also take place as the conversion
increases and the most probable hydrogen acceptor is decreased (i.e. carbonyl groups).
Moreover, we cannot exclude at this stage that the broad band at about 3400 cm-1
could be
also attributed to a slight but effective hydration of the sample. In any case, the contribution
of the OH groups in getting higher Tg values compared to the paraffinic counterparts results
crucial since GPC measurements of all polymer systems showed no significant differences
between their molecular weights (Table 5.3).
Table 5.3 GPC measurements of the polymers of the A-PPy-OHai series and of A-PPy-CH380.
Experiments Mn (x103) Mw (x10
3) PDI
A-PPy-OH20 3.2 7.6 2.4
A-PPy-OH40 3.0 7.0 2.3
A-PPy-OH60 2.9 6.7 2.3
A-PPy-OH80 2.7 6.2 2.3
A-PPy-CH380 3.4 6.1 1.8
87
A-PPy-OH40/MWCNTs composites
Polymer nanocomposites were designed with the aim to target a flexible, light and
temperature-responsive system. In this attempt, A-PPy-OH40 was chosen among the A-PPy-
OHai polymers because its Tg is close to room temperature, and this is expected to provide
the desirable flexibility of the composite at easily accesible temperatures. Moreover, this
feature supports the use of the A-PPy-OH40/MWCNTs system as a resistive sensor for
temperature variations close to the physiological regime. The A-PPy-OH40/MWCNT
composites were prepared by mixing A-PPy-OH40 with different amounts of MWCNTs
(Table 5.2) in N-methylpyrrolidone. The presence of MWCNTs causes an increase in the Tg
of the composite materials compared to the parent polymer (Figure 5.6).
Figure 5.6 Tg of the composites obtained by mixing A-PPy-OH40 with different wt.% of MWCNTs.
The Tg values were measured by DSC in three temperature cycles from -20 to 180 oC.
According to the literature, the increase of Tg values with MWCNTs loading in polymeric
systems is found to be related to the increment in their viscosity [36, 37] due to the interfacial
interaction between matrix and filler at molecular level. Functional aromatic groups included
in the backbone or as pendant groups in polymer chains, get in contact with the graphitic
surface of the filler via supramolecular π-π interactions. This behaviour hinders the mobility
of the polymer chains and hence increases the Tg [38-40]. Conductive polypyrroles have been
reported as good dispersant agents for CNTs due to the interfacial connection of the filler
with the pyrrole groups via π-π interactions. In our case, the backbone of the polymer
0 2 4 6 8 1010
20
30
40
50
Tg (
oC
)
MWCNT (wt%)
88
contains 36% of pyrrole units that possibly promote effective interations with the graphitic
structure of CNT. In order to determine MWCNTs exfoliation, the 4 wt.% MWCNT/A-PPy-
OH40 nanocomposite was analysed by SEM (Figure 5.7).
Figure 5.7 SEM morphological study of 4 wt.% MWCNT / A-PPy-OH40 nanocomposite.
The SEM image clearly shows the single (unbundled) nanotubes as the dominant species,
thus confirming the good interactions between MWCNTs and the A-PPy-OH40 matrix.
With the aim to further improve the dispersion of the MWCNTs in the polymer matrix in
terms of a fibrous percolative structure, the rubber-like composite containing 4 wt.% of
MWCNT was subjected to thermal annealing at a constant temperature of 100 °C for 3 h and
then cooled down to r.t. During the annealing, in-situ resistivity on the compressed mold
specimen were measured in order to evaluate the effect of the thermal treatment on the
formation of the conductive MWCNTs network within the polymer matrix. Before annealing,
the resistivity of the polymer was infinite but it decreased sharply to ≈ 60 µ·m during the
first minutes of thermal annealing and well persisted after cooling (Figure 5.8).
89
Figure 5.8 In-situ electrical measurement of 4 wt.% MWCNT / A-PPy-OH40 composite during
thermal annealing at 100 oC. In the inset, a magnification of the first annealing instants.
This behaviour indicates that the thermal treatment at 100 °C effectively reorganizes
MWCNTs structures within the polymer bulk in a well-defined percolative network. These
remarkable changes easily occurred upon a mild thermal treatment, which is however
effective in promoting polymer matrix mobility.
A morphological study of the composite analyzed by SEM in contrast mode [32] corroborates
the electrical measurements of the nanocomposite before and after annealing (Figure 5.9).
The micrographs taken from the surface of freshly teared samples revealed that bundles of
MWCNTs are still present in the A-PPy-OH40 before annealing as evidenced at high
magnification (Figure 5.9B). Conversely, after annealing a fiber-like polymer/MWCNT
network is observed (Figures 5.9C). This indicates that the thermal treatment at 100 °C favors
MWCNTs debundling and promotes their homogeneus dispersion in the polymer matrix
(Figure 5.9D) in a percolative network.
0 20000 40000 60000 800001E-5
1E-4
1E-3
0,01 Resistivity
Temperature
Time (sec)
Res
isti
vit
y (
Oh
m.m
)
0
20
40
60
80
100
Tem
peratu
re oC
0 2000 4000 6000 8000 10000 12000 14000 16000
1E-4
1E-3
0,01
Res
isti
vit
y (
Oh
m.m
)
Time (sec)
90
Figure 5.9 SEM morphological study of 4 wt.% MWCNT / A-PPy-OH40 composite before (A, B)
and after thermal annealing (C, D) at 100 oC.
The thermal history of the 4 wt.% MWCNT / A-PPy-OH40 nanocomposite after thermal
annealing was evaluated by DSC analysis in order to establish the resilience of the material
under the investigated temperature regime (Figure 5.10).
Figure 5.10 Three consecutive DSC thermal cycles (indicated by the arrow) of 4 wt.% MWCNTs / A-
PPy-OH40 composite after annealing.
-20 0 20 40 60 80 100 120 140 160 180
-2
0
2
Hea
t fl
ow
(W
/g)
end
o u
p
Temperature (oC)
91
The similarity in each consecutive thermal cycle demonstrates the resilient character of the
material without any sign of thermal degradation until 180 °C. The thermal traces also
indicate that the same thermodynamic response of the composite remains upon heating, thus
indicating no phase separation between the components. Considering that the Tg of the
composite is around 40 °C, it is worth noting that the sample does not display any
endothermic transition during the annealing at 100 °C. This is another prove of the role
played by the –OH groups in keeping the dimension stability of the composite. Figure 5.11
shows a picture with two different materials after being subjected to dynamic mechanical
thermal analysis (DMTA) (from r.t. to 35 °C).
Figure 5.11 Pictures of two different nanocomposites displaying their differences in dimension
stability (damping capability of the material) after being tested by dynamic mechanical thermal
analysis DMTA (Data about modulus not showed for brevity).
A-PPy-OH40/MWCNT is capable to keep its dimensions. Contrarily, A-PPy-
CH380/MWCNT gets completely deformed. This suggests that the presence of hydrogen
bonding plays a relevant role in giving dimensional stability to the material. The resistivity
response of the polymer/MWCNT nanocomposite towards consecutive temperature cycles
was monitored between r.t. and 100 °C (Figure 5.12).
Figure 5.12 In-situ electrical measurements of 4 wt.% MWCNT / A-PPy-OH40 composite during
three consecutive thermal cycles between r.t. and 100 oC.
4 wt.% MWCNT / A-PPy-OH40 composite 4 wt.% MWCNT / A-PPy-CH380 composite
0 3000 6000 9000 12000 15000 180005E-5
5.1E-5
5.2E-5
5.3E-5
5.4E-5
5.5E-5
5.6E-5
5.7E-5
5.8E-5
5.9E-5
6E-5
6.1E-5
6.2E-5 Resistivity
Temperature
Time (sec)
Res
isti
vit
y (
Oh
m.m
)
120
100
80
60
40
20
0
Tem
peratu
re oC
92
Reproducible resistance variations were observed, with maximum amplitudes of 51 to 59
µOhm·m within the temperature interval of 80 °C. Moreover, the resistivity–temperature
profile was very reproducible and with a negative temperature, thus proving that the
percolation network does not experience any significant changes during the heating-cooling
cycles. The electrical response with temperature shows a negative temperature coefficient of
resistance [41, 42] of -0.002 K-1
(average of three temperature cycles), an absolute value that
is comparable to the highest values found in metals (0.0037–0.006 K-1
, 0.00385 K-1
for a
Pt100 sensor) and similar to other CNT/polymer nanocomposites reported in the literature [6,
11, 13].
5.4 Conclusions
We have demonstrated that a thermoplastic pyrrole-containing matrix is an effective
dispersant for MWCNTs exfoliation. The polymer was prepared via the Paal-Knorr
modification of an alternating aliphatic polyketone (PK) with OH-amine compound. The
chemical reaction turns the waxy PK into a flexible rubber-like OH-functionalized pyrrole-
containing polymer with tuneable Tg depending on the amounts of -OH groups. The polymer
is able to generate non-covalent functionalization of the MWCNT graphitic materials through
effective π-π interactions, and that the exfoliation process does not do significant damage to
the one-dimensional CNT structure. SEM micrographs and DSC traces demonstrate that the
A-PPy-OH/MWCNTs nanocomposite is capable to undergo continuous thermal cycles from -
20 to 180 oC without any sign of interphase modification and matrix degradation. Notably,
the conductive CNT network is maintained after several temperature cycles (from r.t. to 100
oC) proving the remarkable stability of the MWCNT homogeneous dispersion within the
polymer matrix. Measurements repeated over three successive heating cycles revealed highly
reproducible resistivity variations with negative temperature coefficient of about -0.002 K-1
,
an absolute value comparable to the values found in metals. Overall, this data consistently
support the use of A-PPy-OH/MWCNT nanocomposite as a soft and highly reproducible
resistive sensor for temperature variations.
5.5 References
1. McNally, Tony & Pötschke, Petra. Polymer-carbon nanotube composites. Philadelphia,
USA: Woodhead Publishing Limited, 2011.
2. IIJIMA S. Helical Microtubules of Graphitic Carbon. Nature 1991;354(6348):56-58.
3. Moniruzzaman M, Winey KI. Polymer nanocomposites containing carbon nanotubes.
Macromolecules 2006;39(16):5194-5205.
93
4. Wei B, Vajtai R, Ajayan P. Reliability and current carrying capacity of carbon nanotubes.
Appl Phys Lett 2001;79(8):1172-1174.
5. Pop E, Mann D, Wang Q, Goodson K, Dai H. Thermal conductance of an individual
single-wall carbon nanotube above room temperature. Nano Letters 2006;6(1):96-100.
6. Matzeu G, Pucci A, Savi S, Romanelli M, Di Francesco F. A temperature sensor based on
a MWCNT/SEBS nanocomposite. Sensors and Actuators A-Physical 2012;178:94-99.
7. De Volder MFL, Tawfick SH, Baughman RH, Hart AJ. Carbon Nanotubes: Present and
Future Commercial Applications. Science 2013;339(6119):535-539.
8. Kin-Tak Lau A. Carbon Nanotube-Reinforced Nanocomposites. In: Leng J, Kin-Tak Lau
A, editors. : CRC Press, 2011. p. 19.
9. Biver T, Criscitiello F, Di Francesco F, Minichino M, Swager T, Pucci A.
MWCNT/perylene bisimide water dispersions for miniaturized temperature sensors. RSC
Advances 2015;5(80):65023-65029.
10. Giuliani A, Placidi M, Di Francesco F, Pucci A. A new polystyrene-based
ionomer/MWCNT nanocomposite for wearable skin temperature sensors. Reactive &
Functional Polymers 2014;76:57-62.
11. Calisi N, Giuliani A, Alderighi M, Schnorr JM, Swager TM, Di Francesco F, Pucci A.
Factors affecting the dispersion of MWCNTs in electrically conducting SEBS
nanocomposites. European Polymer Journal 2013;49(6):1471-1478.
12. Calisi N, Giuliani A, Alderighi M, Schnorr JM, Swager TM, Di Francesco F, Pucci A.
Factors affecting the dispersion of MWCNTs in electrically conducting SEBS
nanocomposites. European Polymer Journal 2013;49(6):1471-1478.
13. Neitzert HC, Vertuccio L, Sorrentino A. Epoxy/MWCNT Composite as Temperature
Sensor and Electrical Heating Element. Ieee Transactions on Nanotechnology
2011;10(4):688-693.
14. Li C, Thostenson ET, Chou T. Sensors and actuators based on carbon nanotubes and their
composites: A review. Composites Sci Technol 2008;68(6):1227-1249.
15. Majumder A, Khazaee M, Opitz J, Beyer E, Baraban L, Cuniberti G. Bio-
functionalization of multi-walled carbon nanotubes. Physical Chemistry Chemical Physics
2013;15(40):17158-17164.
16. McNally, Tony & Pötschke, Petra. Preparation and processing of polymer-carbon
nanotube composites. In: Cambridge, UK: Woodhead Publishing Limited, 2011. p. 3.
17. Zydziak N, Huebner C, Bruns M, Barner-Kowollik C. One-Step Functionalization of
Single-Walled Carbon Nanotubes (SWCNTs) with Cyclopentadienyl-Capped
Macromolecules via Diels-Alder Chemistry. Macromolecules 2011;44(9):3374-3380.
18. Chang C, Liu Y. Functionalization of multi-walled carbon nanotubes with furan and
maleimide compounds through Diels-Alder cycloaddition. Carbon 2009;47(13):3041-3049.
19. Guadagno L, De Vivo B, Di Bartolomeo A, Lamberti P, Sorrentino A, Tucci V, Vertuccio
L, Vittoria V. Effect of functionalization on the thermo-mechanical and electrical behavior of
multi-wall carbon nanotube/epoxy composites. Carbon 2011;49(6):1919-1930.
20. Sahoo NG, Rana S, Cho JW, Li L, Chan SH. Polymer nanocomposites based on
functionalized carbon nanotubes. Progress in Polymer Science 2010;35(7):837-867.
21. Chen G, Shaffer M, Coleby D, Dixon G, Zhou W, Fray D, Windle A. Carbon nanotube
and polypyrrole composites: Coating and doping. Adv Mater 2000;12(7):522-+.
22. Su Y, Zhitomirsky I. Asymmetric electrochemical supercapacitor, based on polypyrrole
coated carbon nanotube electrodes. Appl Energy 2015;153:48-55.
23. Hodge IM. Method of enhancing the flexibility of polypyrrole structures 1987.
24. Lim TH, Oh KW, Kim SH. Polypyrrole/MWCNT-gr-PSSA Composite for Flexible and
Highly Conductive Transparent Film. J Appl Polym Sci 2012;123(1):388-397.
94
25. Drent E. Process for the preparation of polyketones 1984(EP 0121965 A2).
26. Zhang Y, Broekhuis AA, Stuart MCA, Picchioni F. Polymeric Amines by chemical
modifications of alternating aliphatic polyketones. J Appl Polym Sci 2008;107(1):262-271.
27. Mul WP, Dirkzwager H, Broekhuis AA, Heeres HJ, van der Linden AJ, Orpen AG.
Highly active, recyclable catalyst for the manufacture of viscous, low molecular weight, CO-
ethene-propene-based polyketone, base component for a new class of resins. Inorg Chim Acta
2002;327:147-159.
28. Toncelli C, De Reus DC, Picchioni F, Broekhuis AA. Properties of Reversible Diels-
Alder Furan/Maleimide Polymer Networks as Function of Crosslink Density.
Macromolecular Chemistry and Physics 2012;213(2):157-165.
29. Zhang Y, Broekhuis AA, Picchioni F. Thermally Self-Healing Polymeric Materials: The
Next Step to Recycling Thermoset Polymers?. Macromolecules 2009;42(6):1906-1912.
30. Tee BC-, Wang C, Allen R, Bao Z. An electrically and mechanically self-healing
composite with pressure- and flexion-sensitive properties for electronic skin applications.
Nature Nanotechnology 2012;7(12):825-832.
31. Loos J, Alexeev A, Grossiord N, Koning CE, Regev O. Visualization of single-wall
carbon nanotube (SWNT) networks in conductive polystyrene nanocomposites by charge
contrast imaging. Ultramicroscopy 2005;104(2):160-167.
32. Deng H, Skipa T, Bilotti E, Zhang R, Lellinger D, Mezzo L, Fu Q, Alig I, Peijs T.
Preparation of High-Performance Conductive Polymer Fibers through Morphological Control
of Networks Formed by Nanofillers. Advanced Functional Materials 2010;20(9):1424-1432.
33. Silverstein RM, Bassler GC, Morrill TC. Spectrometric identification of organic
compounds. Canada: John Wiley and sons, Inc., 1991.
34. Huang Y, Unni AK, Thadani AN, Rawal VH. Hydrogen bonding: Single enantiomers
from a chiral-alcohol catalyst. Nature 2003;424(6945):146-146.
35. Seebach D, Beck A, Heckel A. TADDOLs, their derivatives, and TADDOL analogues:
Versatile chiral auxiliaries. Angewandte Chemie-International Edition 2001;40(1):92-138.
36. Wu D, Wu L, Zhang M. Rheology of multi-walled carbon nanotube/poly(butylene
terephthalate) composites. Journal of Polymer Science Part B-Polymer Physics
2007;45(16):2239-2251.
37. Du F, Scogna R, Zhou W, Brand S, Fischer J, Winey K. Nanotube networks in polymer
nanocomposites: Rheology and electrical conductivity. Macromolecules 2004;37(24):9048-
9055.
38. Prado LASdA, Kwiatkowska M, Funari SS, Roslaniec Z, Broza G, Schulte K. Studies on
Morphology and Interphase of Poly(butylene terephthalate)/Carbon Nanotubes
Nanocomposites. Polym Eng Sci 2010;50(8):1571-1576.
39. Meuer S, Braun L, Schilling T, Zentel R. alpha-Pyrene polymer functionalized
multiwalled carbon nanotubes: Solubility, stability and depletion phenomena. Polymer
2009;50(1):154-160.
40. Lee J, Yoon S, Kim KK, Cha I, Park YJ, Choi J, Lee YH, Paik U. Exfoliation of single-
walled carbon nanotubes induced by the structural effect of perylene derivatives and their
optoelectronic properties. Journal of Physical Chemistry C 2008;112(39):15267-15273.
41. Naeemi A, Meindl JD. Physical modeling of temperature coefficient of resistance for
single- and multi-wall carbon nanotube interconnects. IEEE Electron Device Lett
2007;28(2):135-138.
42. Vollebregt S, Banerjee S, Beenakker K, Ishihara R. Size-Dependent Effects on the
Temperature Coefficient of Resistance of Carbon Nanotube Vias. IEEE Trans Electron
Devices 2013;60(12):4085-4089.
95
6 Electrically-responsive thermoset nanocomposite based on Diels-Alder chemistry
Abstract. In this work, a nanocomposite-based thermoset material with thermally reversible
cross-linkages was designed and prepared by cross-linking a furan-functionalized polyketone
(PK-Fu) with bis-maleimide and reinforced with multi-walled carbon nanotubes (MWCNTs)
via Diels-Alder (DA) reversible cycloaddition. The main novel property of this material is
that self-healing can be induced by resistive heating. The incorporation of 5 wt.% of
MWCNTs results in an increased modulus of the material and makes it thermally and
electrically conductive. XPS analysis indicates that the MWCNTs, due to their
diene/dienophile character, covalently interact with both furan and maleimide groups via
Diels-Alder reaction, leading to good interfacial adhesion between filler and matrix.
Moreover, the softening point (taken as tangent δ in DMTA measurements) increases to 155
°C upon the addition of MWCNTs, as compared to the value of 130 °C for the unreinforced
thermoset. After thermo-mechanical strain is applied, the composite diminishes its
mechanical strength since heating above 150 °C triggers the retro-DA process that disrupts
the network. Nevertheless, the mechanical properties of the composite are completely
recovered by reconnecting the decoupled DA linkages via electrical resistive heating.
Keywords: Self-healing, resistive heating, conductive thermoset nanocomposite,
reworkability, reversible Diels-Alder cycloaddition.
96
6.1 Introduction
Thermoset polymers are an important class of materials used in a broad range of applications,
including coatings, adhesives and electrical insulators [1]. Thermosets, unlike thermoplastics,
are characterized by a curing reaction that transforms two liquid components, or a paste [2],
into a solid network structure. Due to this network, thermosets offer many superior properties
compared to thermoplastics, such as mechanical strength, dimensional stability at elevated
temperature and solvent resistance. However, thermoset systems are still facing challenges
related to thermal, chemical and photo-degradation and high brittleness. Typically,
undetectable micro-cracks can appear under loading thus leading to macroscopic fractures at
a relatively small strain [3-5]. To overcome this drawback, thermosets have been combined
with different kinds of fillers such as silica and glass beads [6, 7], engineered thermoplastics
as polyphenylene oxide [8, 9], rubbery core-shell particles [10-12], and carbon fibers [13, 14]
in order to design suitable load-bearing materials for structural applications.
Intensive research efforts are currently dedicated to synthesize thermoset nanocomposites
with the aim of improving strength, modulus, and toughness of polymer matrices [15-18]. As
a common procedure, the weight content of the filler in the matrix is tailored in order to gain
the maximal reinforcement from the filler considering its aspect ratio and loading.
Accordingly, the inclusion of small amounts of filler (< 5 wt.%) is often used to improve the
mechanical properties of the matrix. Contrarily, filler amounts exceeding the loading
threshold possibly generate the failure of the whole system due to the loss of effective
interactions between the components at the interface [19, 20]. In this respect, the chemical
functionalization of nanofillers via covalent attachment of compatibilizing functional groups
or polymers has been reported to increase the mechanical performance of the nanocomposites
even using relatively high filler loading [20-22].
A promising strategy employed in the fabrication of long-lasting nanocomposite systems is
the incorporation of healing mechanisms for damage repair [23, 24]. Among several
approaches, self-mendable thermoset nanocomposites have been designed according to
different procedures in order to create autonomous and non-autonomous self-repairing
systems [25-28]. In the case of non-autonomous self-mendable systems (i.e. intrinsic self-
healing [28, 29]), the matrix experiences healing damage thanks to the presence of active
functional moieties, which are triggered by external stimuli like heat [30] or light [31]. The
two most exploited thermal methods currently used for the healing process are based on
97
Diels-Alder (DA) chemistry (reversible covalent cycloaddition) and/or hydrogen bonding
(reversible non-covalent interaction)[32-36].
Electrically-induced self-healing is a relatively new concept currently employed in the design
of long-lasting nanocomposites for electronic applications and actuators [37-42]. For these
particular systems, the healing process occurs via the nanoscopic heat generation when the
electrical current passes through a conductive nanostructured network [43]. The so-called
Joule-effect (or resistive heating) activates the intrinsic self-healing ability of thermally self-
mendable matrices in order to heal damage on local areas. Among several advantages, this
approach guarantees a faster healing process even at a nanoscale level, thus representing an
extraordinary alternative for curing and self-healing in engineering thermoset applications
that are impossible for conventional heating procedures [44, 45].
In previous works, alternating aliphatic polyketones (PK) obtained by the copolymerization
of carbon monoxide, ethylene, and propylene were chemically modified by the Paal-Knorr
reaction to introduce furan groups directly attached to the backbone chain. The chemical
reaction proceeded in the bulk with high yields and relatively fast kinetics, producing water
as the only by-product [46-48]. The grafted furan groups allowed the formation of three-
dimensional polymer network structures after being cross-linked with aromatic bis-
maleimide. The materials indeed formed thermally reversible and self-healing thermosets by
means of DA and retro-DA (r-DA) sequence employing conventional heating procedures.
Herein we report on the intrinsic self-healing ability induced by resistive heating of an
electrically conductive thermoset nanocomposite based on the chemical modification of an
alternating aliphatic polyketone grafted with furan groups (PK-Fu) via Paal-Knorr reaction,
cross-linked with bis-maleimide and reinforced with MWCNTs via reversible Diels-Alder
cycloaddition. Spectroscopic investigations were used to assess the modification of the
starting polyketone, the cross-linking of the resulting polymer, and the chemical bonding with
the nanofiller via DA reaction. Thermo-mechanical tests allowed evaluating the mechanical
performance and self-healing ability of the composite. Results concerning the material
modulus were correlated with electrical measurements, spectroscopy and IR thermography.
The morphology of the system was analysed by electronic microscopy before and after
healing by resistive heating.
98
6.2 Experimental section
Reagents
The alternating aliphatic polyketone (PK30, MW 2687 Da) presents a total olefin content of
30% of ethylene and 70% of propylene [49]. Furfurylamine (Fu) (Aldrich, ≥ 99 %) was
freshly distilled before used, and benzylamine (Bea) (Sigma Aldrich 99 %), multi-walled
carbon nanotubes (MWCNTs) (O.D. 6-9 nm, average length 5 µm, Sigma-Aldrich 95 %
carbon), DMSO-d6 (Laboratory-Scan, 99.5%), (1,1-(methylenedi-4,1-phenylene)bis-
maleimide (b-Ma) (Sigma Aldrich 95 %), 1-Methyl-2-pyrrolidone (NMP, Sigma-Aldrich,
99.5 %), tetrahydrofuran (THF, Laboratory-Scan, 99.5 %), chloroform (CHCl3, Laboratory-
Scan, 99.5 %) and deuterated chloroform (CDCl3, Sigma Aldrich 99.8 atom% D) were
purchased and used as received.
Functionalization of Polyketone with furan and benzyl groups
The reaction between PK and furfurylamine was carried out in bulk by the Paal-Knorr
reaction [50] and the molar ratio between the reactants (i.e. the di-carbonyl group of
polyketone and the amine group of furfurylamine) was established as percentages with a
maximal conversion of 80 % according to Zhang et al. 2009 and Toncelli et al. 2012 [47, 48].
In order to establish more clearly the role of the furan motifs, a polymer grafted with benzyl
groups was prepared by the Paal-Knorr reaction and used as a reference. This so-called PK-
Bea displays the same backbone structure of PK-Fu but non-reactive pendant groups, at least
in our conditions, by Diels-Alder cycloaddition (Figure 6.1, see Paal-Knorr experimental
procedures in previous chapters).
Figure 6.1 Schematic representation of PK functionalized with furan (PK-Fu) or benzyl (PK-Bea)
groups via Paal-Knorr reaction.
99
PK-Fu Cross-linked with b-Ma via Diels-Alder reaction
The DA reaction of PK-Fu (10 g) with bis-maleimide was carried out at a furan / maleimide
molar ratio of 1:1 using THF as solvent (≈ 10 wt.% polymer based on solvent) in a 250 mL
round-bottom flask equipped with a magnetic stirrer and a reflux condenser. The reaction
mixture was heated up to 50 °C for 24 h to form the polymer network. After reaction, the
cross-linked polymer was dried at 50 °C under vacuum overnight to remove the solvent.
Functionalization of MWCNTs with PK-Fu or b-Ma via Diels-Alder reaction
PK-Fu (0.95 g) was dissolved in 5 mL of NMP and added to 0.05 g of MWCNT (previously
sonicated for 30 min in 5 mL of NMP). The reaction was set under vigorous stirring at 50 °C
for 24 h using an oil bath equipped with a temperature controller. The reaction mixture was
repeatedly washed with THF, filtered and the remaining solvent removed under vacuum at 50
°C for 48 h. The same procedure was used to mix MWCNTs with bis-maleimide. A sample
containing MWCNTs and PK30-Bea was used as a reference since no chemical interaction is
expected between the components. The percentage of grafted product (grafted (%)) can be
calculated as follows:
𝑔𝑟𝑎𝑓𝑡𝑒𝑑 (%) = 𝑃𝑟𝑜𝑑𝑟𝑒𝑐𝑜𝑣
𝐺𝑟𝑎𝑓𝑝𝑟𝑜𝑑∙ 100 (1)
where Prodrecov represents the amount of material recovered after filtering the excess of
compound that did not react with MWCNTs; Grafprod represents the amount of PK-Fu, PK-
Bea or b-Ma (in grams) grafted on the MWCNTs surface. The Grafprod is calculated as
follows:
𝐺𝑟𝑎𝑓𝑝𝑟𝑜𝑑 = 𝑀𝑤𝑦 ∙ 𝑚𝑜𝑙𝑒𝑠 (𝑁) (2)
where Mw y
represents the molecular weight of the functionalized pyrrolic unit of PK-Fu and
PK-Bea (in the case of b-Ma, the two nitrogens of the molecule are considered) and moles
(N) are the moles of the pyrrolic unit or maleimide groups according to the moles of nitrogen
obtained by elemental analysis (see Figure 6.1). The moles of nitrogen are calculated as
follows:
𝑚𝑜𝑙𝑒𝑠 (𝑁) =𝑔(𝑁)
𝐴𝑚(𝑁) (3)
100
where 𝑔(𝑁) represents the grams of nitrogen according to elemental analysis and Am(N) the
atomic mass of nitrogen. Finally, the 𝑔(𝑁) can be calculated as follows:
𝑔(𝑁) = 𝑃𝑟𝑜𝑑𝑟𝑒𝑐𝑜𝑣 ∙ 𝑁𝑐𝑜𝑛𝑡𝑒𝑛𝑡 (4)
where 𝑁𝑐𝑜𝑛𝑡𝑒𝑛𝑡 represents the percentage of nitrogen estimated by elemental analysis.
PK-Fu / b-Ma / MWCNTs composite
The thermoset nanocomposite was prepared by one-pot solvent-mix containing equimolar
amounts of PK-Fu and b-Ma (at a furan / maleimide ratio of 1:1) and 5 wt.% of MWCNTs.
The reactants were previously dissolved in THF (≈ 10 wt.%) and bath sonicated for 30 min in
a 150 mL round-bottomed flask equipped with a magnetic stirrer. The reaction mixture was
heated up to 50 °C for 24 h to form the cross-linked network under reflux. After reaction, the
solvent was removed under vacuum at 50 °C overnight. The resulting powder was divided
into small pieces of nearly 500 mg that were molded into rectangular bars at 150 °C for 30
min under a pressure of 40 bar. A neat thermoset sample without MWCNTs was also
prepared for comparison. After molding, the samples were cooled down to room temperature
(30 minutes) and then stored at -17 °C for further analysis.
Characterization
The elemental composition of the samples was analyzed using an Euro EA elemental
analyzer. 1H-NMR spectra were recorded on a Varian Mercury Plus 400 MHz apparatus
using deuterated chloroform as solvent. FT-IR spectra were collected using a Perkin-Elmer
Spectrum 2000. The sample pellets were prepared by mixing potassium bromide (KBr) with
the polymer (≈1.5 wt.%). The powder was then kept under vacuum at 50 °C for 24 h to
remove residual water. ATR-FTIR spectra were recorded using a Thermo Nicolet NEXUS
670 FT-IR. Differential Scanning Calorimetry (DSC) analysis was performed on a TA-
Instrument DSC 2920 under N2 atmosphere. The samples were weighed (10-17 mg) in an
aluminum pan, which was then sealed. Hereafter, the samples were heated from 0 to 180 °C
and then cooled to 0 °C. Four heating-cooling cycles were performed at a rate of 10 °C/min.
GPC measurements were performed with a HP1100 Hewlett-Packard. The equipment
consists of three 300 x 7.5 mm PLgel 3 µm MIXED-E columns in series and a GBC LC 1240
IR detector. The samples were dissolved in THF (1 mg/mL) and eluted at a flow rate of 1
mL/min at a temperature of 40 °C. The calibration curve was made using polystyrene as
101
standard and the data were interpolated using the PSS WinGPC software. Thermogravimetric
analyses (TGA) were carried out in a nitrogen environment on a PerkinElmer TGA 7
instrument from 20 °C to 700 °C at a heating rate of 10 °C/min. Dynamic Mechanical
Thermal Analyses (DMTA) were conducted on a rheometrics scientific solid analyzer (RSA
II) under air environment using the dual cantilever mode at an oscillation frequency of 1 Hz
and a heating rate of 3 °C/min. The samples for DMTA analysis were prepared by
compression molding of 500 mg of the composite into rectangular bars (6 mm wide, 1 mm
thick, 54 mm long) at 150 °C for 30 min under a pressure of 40 bar to ensure full
homogeneity. Electrical measurements were performed on the rectangular bars used for
DMTA analysis. The setup consisted of a power supply (EA-PS 3150-04 B) and a multimeter
(FLUKE 175). Electrical parameters were measured on samples connected to a conventional
circuit using copper clamps holders. The surface resistivity (ρs) was calculated as follows:
ρs = (𝑉/𝐿)/(𝐼/𝑊) (5)
where V is the voltage supplied in the electrical circuit, L is the distance in meters between
the electrodes connected to the sample, I is the current measured in amperes and W is the
width of the sample in meters. Thermal images were obtained using a FLUKE IR
thermometer camera (VT02) from samples subjected to electrical current. X-ray
photoelectron spectroscopy (XPS) was carried out with a SSX-100 (Surface Science
Instruments) spectrometer equipped with a monochromatic Al Kα X-ray source (hν = 1486.6
eV) that operates at a base pressure of 3 × 10−10
mbar. The energy resolution was set at 1.45
eV, the photoelectron take-off angle was 37° with respect to the surface normal and the
diameter of the analyzed spot was 600 μm. Spectra were collected at a minimum of two
different spots on each sample, checked for consistency and averaged for each spectral
interval. The data were fitted using the Winspec software [51] applying the Shirley
background and a linear combination of Gaussian and Lorentzian peaks (mixing ratio of 0.8).
All samples were re-suspended in toluene and deposited dropwise on Au substrates. After the
solvent was evaporated, the samples were transferred into ultra-high vacuum via a load-lock
system. Samples containing the pure polymer display broadened peaks due to sample
charging effects. Therefore, the fitting width was set at 2.1 eV. The binding energy scale was
corrected by using the C=C sp2 signal at 284.4 eV as a reference [52]. Scanning electron
microscope micrographs were taken with a Philips XL30S Environmental SEM-FEG
instrument. High resolution images were acquired on freshly broken surfaces.
102
6.3 Results and discussion
PK functionalized with furan and benzyl groups via Paal-Knorr reaction
The solvent-free Paal-Knorr reaction between PK and the amine compounds was carried out
using different molar ratios between the 1,4-di-carbonyl groups of polyketone and
furfurylamine or benzylamine aiming at a maximal conversion of 80%. The summary of the
experimental results is displayed in Table 1.
Table 6.1 Experimental results of PK modified with furfurylamine (PK-Fu) and benzylamine (PK-
Bea).
Run CCO (%)a
η (%)b Tg (
oC)
c PDI
d
PK-Fu 74 92 31 2.3
PK-Bea 73 91 42 2.2
a Carbonyl conversion,
b conversion efficiency.
c glass transition temperature,
d polydispersity index.
Notably, functionalized polyketones display a carbonyl conversion (CCO, measured by
elemental analysis) close to the target conversion (efficiency, η > 90 %, see Table 6.1). This
result validates the robustness and versatility of the Paal-Knorr reaction with polyketones.
For brevity, only the spectral characterization of PK-Fu is displayed in Figure 6.2. Figure
6.2A shows the 1H-NMR spectrum of the polymer mixed with furfurylamine before and after
the chemical reaction. After modification, the characteristic peaks of the pyrrole appear in the
spectrum, namely the proton signal ascribed to CH2 groups between the pyrrole and the furan
groups at 4.9 ppm, the one related to the pyrrole group at 5.8 ppm and proton signals of the
furan moieties at 5.9, 6.2 and 7.3 ppm [46, 47]. The FT-IR spectrum of the same sample is
displayed in Figure 6.2B. It is possible to notice the appearance of C-H stretching related to
the heterocyclic groups around 3150-3115 cm-1
(pyrrole and furan groups), the C=O
stretching of the residual carbonyl groups at 1707 cm-1
, the C=C stretching of the
heterocyclic groups at 1507 cm-1
, the pyrrole C-N stretching at 1345 cm-1
, the furan C-O-C
stretching at 1073 cm-1
and eventually the out-of-plane bending of the furan ring C-H bonds
at 735 cm-1
.
103
Figure 6.2 A) 1H-NMR spectra of PK before (b) and after chemical modification with furfurylamine
(a). B) FT-IR spectra of PK after chemical modification with furfurylamine a) Vs C-H heterocyclic
groups (3150-3115 cm-1
), b) Vs C=O 1707 cm-1
, c) Vs C=C 1507 cm-1
, d) Vs N-C 1345 cm-1
, e) Vs C-
O-C 1073 cm-1
, f) cyclic out-of-plane C-H bending 735 cm-1
[46].
The Tg of the functionalized polymers was measured by DSC (Table 6.1). As expected, the
chemical modification of the di-carbonyl arrangement into pyrrole groups increases the
rigidity of the backbone. Indeed PK30-Fu and PK30-Bea show a Tg of 31 °C and 42 °C,
respectively, i.e. much higher than that of PK before modification (-12 °C). GPC
measurements showed no significant differences in the PDI of the two systems, indicating the
absence of significant irreversible side reactions. The comparison of the two systems suggests
that benzyl pendant groups confer stronger supramolecular interaction among the
macromolecular chains with respect to that provided by furan moieties. This highlights the
versatility of the PK system, whose modification with different amino compounds grafted on
the same backbone allows tuning the Tg value of the ultimate polymer.
3000 2500 2000 1500 1000
Tra
nsm
itta
nce
(a.
u.)
Wavenumber (cm -1)
8 7 6 5 4 3 2 1 0
PPM
104
Functionalization of MWCNTs with PK-Fu or b-Ma via Diels-Alder reaction
Figure 6.3 displays schematically the reaction between MWCNTs and PK-Fu or b-Ma. The
amount of PK-Fu and b-Ma attached to the MWCNTs surface was evaluated by elemental
analysis and TGA (Table 6.2).
Figure 6.3 Functionalization of MWCNTs with A) PK-Fu (diene) and B) bis-maleimide (dienophile)
via Diels-Alder reaction.
Table 6.2 Experimental results for the functionalization of MWCNTs with PK-Fu, PK-Bea and bis-
maleimide.
Run N content (%) Grafted (%)a Grafted (%)
b
MWCNT/PK30-Fu80 2.45 34 27
MWCNT/PK30-Bea80 0.83 13 9
MWCNT/b-Ma 0.62 8 6
Grafted compounds (%) on the surface of MWCNTs are estimated by a
elemental analysis (nitrogen
(N) content) and b TGA.
The results obtained by EA clearly establish the presence of nitrogen in the final samples.
Conversely, no presence of nitrogen was found in un-functionalized MWCNTs within the
detection limit of the apparatus (0.01 wt.%). The values of nitrogen obtained for the grafted
MWCNTs/PK-Fu suggest that considerably higher functionalization of the filler is obtained
as compared to MWCNTs/b-Ma. This is quite logic as PK-Fu contains many nitrogen atoms
105
along the chain. In the case of MWCNTs mixed with PK-Bea, no covalent interaction is
expected, thus the result obtained by EA might only suggest that effective physical mixing
between the polymer and the filler occurs via π-π staking interactions [53-55].
Thermogravimetric analyses were also performed to establish the amount of PK-Fu and bis-
maleimide covalently attached to MWCNTs (Table 6.2 and Figure 6.4). TGA results show
clear differences between the un-functionalized and the functionalized MWCNTs. However,
lower values of grafting are obtained from TGA results, as compared to EA, which might be
explained by the fact that at 700 °C no degradation of the grafted materials is detected (see
the slope at the final temperature of degradation in Figure 6.4). The temperature of
degradation for the functionalized MWCNTs systems starts around 100 °C due to the
presence of residual water. The larger weight loss of MWCNT-PK-Fu clearly indicates the
successful attachment of the polymer on the surface of the MWCNTs. Moreover, the weight
loss fraction of the grafted compounds follows the same trend as compared with the values of
nitrogen calculated from EA (see Table 6.2).
Figure 6.4 TGA analyses of pristine and functionalized MWCNTs with PK-Fu, b-Ma and the
reference sample PK-Bea.
In literature, the formation of covalent bonds between cyclopentadiene, furan and bis-
maleimide with CNTs via DA chemistry has been already proved by means of X-ray
photoelectron spectroscopy [52, 56]. Herein, XPS analysis of the MWCNTs functionalized
with PK-Fu clearly revealed the presence of nitrogen 1s and oxygen 1s signals (Figure 6.5).
100 200 300 400 500 600 70070
75
80
85
90
95
100
Rel
ativ
e w
eig
ht
(%)
Temperature (oC)
a) MWCNTs
b) MWCNTs-b-Ma
c) MWCNTs-PK-Bea
d) MWCNTs-PK-Fu
106
Figure 6.5 X-ray photoemission spectra of the C 1s, N 1s and O 1s core level regions of MWCNTs
functionalized with PK-Fu. Data and fitting lines are shown.
It is worth noting that in the carbon 1s spectra, differences between the modified and un-
modified MWCNTs surfaces can be identified; for the un-modified MWCNTs the main peak
at 284.5 eV, accompanied by the shake-up feature at about 6.0 eV higher binding energy
characteristic of π-π conjugated systems, perfectly fits with C sp2 reported in the literature
[52, 56]. On the other hand, the C 1s spectrum of the pure polymer shows a broad feature
peaked at 285 eV and resulting from three components: C-C sp2, -CH2 -CH3 sp
3 and CH2-C-
CH3, which are too closely spaced in binding energy [52] to be resolved in our experimental
conditions. The spectrum for the pure polymer also shows O 1s components attributed to
O=C at 532.4 eV and to O-C at 533.8 eV, as well as an N 1s peak at 401.5 eV; all these
components perfectly agree with PK-Fu structure. For the product obtained after the grafting
PK-Fu to the MWCNTs, the presence of the polymer is evident from the N 1s and O 1s
intensities. In the C 1s spectrum, a broad peak was observed which was here fitted with a
linear combination of the pure polymer and MWCNTs signal (respectively in dotted and
dashed lines). While it is difficult to conclude from the C 1s spectrum that the sidewall
functionalization of MWCNTs with PK-Fu proceeds via C-C bonding from the C 1s
spectrum, in the O 1s spectrum the O-C component results shifted by 1 eV to higher binding
energy, clearly pointing at a charge redistribution resulting from the furan attachment on the
MWCNTs surface.
PK-Fu / b-Ma / MWCNTs composite
The thermoset nanocomposite was obtained in the form of highly homogeneous rectangular
bars by one-pot solvent-mix containing equimolar amounts of PK-Fu and b-Ma (ratio 1:1
107
between Fu and Ma groups) and 5 wt.% of MWCNTs followed by compression-molding
[46].
The neat thermoset (not shown for brevity) and the nanocomposite (Figure 6.6) were
characterized by DSC to determine their thermal behavior, reversibility and the exo / endo-
thermal processes related to the DA and retro-DA sequence, respectively. The thermograms
display a broad endothermic transition in the range of temperature between 120 and 180 °C
for each consecutive thermal cycle. The similarity between the three thermal cycles clearly
indicates the reworkable character of the cross-linked PK-Fu even when reinforced with the
MWCNTs. The endothermic transition corresponds to the r-DA process. Therefore, the peak
of the curves corresponds to the average temperature at which the majority of the DA adducts
are broken and the area under the curve associated to this peak is related to the energy
absorbed during the cleavage of the DA adducts.
The DSC analysis of the nanocomposite evidences that the presence of MWCNTs has no
apparent effect on the thermal behaviour of the thermoset [57]. Indeed, the energy associated
to the r-DA process was calculated to be 11.1 J/g for the nanocomposite and 10.9 J/g for the
neat thermoset, thus suggesting that the presence of the filler does not interfere with the r-DA
process of the matrix [57]. In other words, even if a covalent interaction between matrix and
filler occurs, the corresponding thermal transition is not visible in the DSC traces due to
either overlap with the one associated to Fu-Ma or simply relatively low concentration.
Figure 6.6 DSC thermal cycles of PK-Fu cross-linked with b-Ma and reinforced with MWCNTs.
0 20 40 60 80 100 120 140 160 180
-4
-2
0
2
Hea
t fl
ow
(W
/g)
Temperature (oC)
108
DMTA analysis performed on rectangular bars of materials shows that the filler leads to clear
enhancements of softening point (peak Tan δ), loss and elastic moduli with respect to the neat
thermoset system (Figure 6.7).
Figure 6.7 Dynamic Mechanical Thermal Analysis of PK-Fu cross-linked with b-Ma (filled squares)
and PK-Fu cross-linked with b-Ma and reinforced with 5 wt. % of MWCNTs (empty squares).
In particular, the softening point (Tan δ) is increased from ~130 °C for the neat thermoset to
~155 °C for the composite. This can be explained by the fact that MWCNTs reinforce the
inner-frame structure of the matrix that consequently increases its rigidity. This effect is aided
by the interfacial interaction between the matrix and the filler via DA cycloaddition, as
suggested by XPS analyses. As a result, the filler helps in dispersing the applied force and
consequently improves the mechanical performance of the composite. Another phenomenon
that possibly contributes in the softening point enhancement is the potential catalytic effect
exerted by MWCNTs on the DA reaction between PK-Fu and b-Ma, which leads to higher
cross-linking density [58].
The thermo-mechanical features of PK-Fu cross-linked with b-Ma and reinforced with
MWCNTs were also determined after compression at 150 °C and 40 bar for 30 min. This
process was supposed to favour r-DA mechanism and the consequent de-cross-linking of the
thermoset nanocomposite. Figure 6.8 displays the thermo-mechanical behaviour of the de-
cross-linked network and after healing by resistive heating at 35 V for 24h.
40 60 80 100 120 140 160
1E7
1E8
1E9
1E10
Temperature (oC)
E'E
" (P
a)
0.00
0.05
0.10
0.15
0.20
Tan
109
Figure 6.8 DMTA of PK-Fu cross-linked with b-Ma and reinforced with 5 wt.% of MWCNTs, after
pressed at 150 °C and 40 bar for 30 min (filled squares) and after healing by passing a current through
the sample during 24 h using 35 V.
It is worth noting that the elastic modulus (E’) and the softening temperature (Tan δ) are
significantly improved after healing as compared with the system after molding. The
electrical current applied as energy source effectively induces heat dissipation from
MWCNTs, which triggers the reconnection between decoupled DA adducts thus re-
establishes the cross-linked network [59]. A schematic representation of the proposed
mechanism is reported in Figure 6.9.
Figure 6.9 Schematic representation of PK-Fu cross-linked with b-Ma and reinforced with
MWCNTs. Arrows display the process of: 1) cross-linking (DA), 2) molding (partial r-DA) and 3)
healing (DA) with resistive heating.
40 60 80 100 120 140 160
1E7
1E8
1E9
1E10
Before healing
After healing
Temperature (oC)
E'E
" (P
a)
0.00
0.05
0.10
0.15
0.20
0.25
Tan
110
In order to confirm this idea, we monitored the temperature of the sample when a current
goes through it (Figure 6.10).
Figure 6.10 (A) Photograph of the PK-Fu/b-Ma/MWCNTs composite sample connected to the
electrical circuit and its thermal images before (B) and during the application of a voltage of 35 V (C).
Notably, the well distributed red colour all along the film surface clearly suggests the
homogeneous distribution of MWCNTs in the thermoset nanocomposite. The temperature
reached by the network during the application of the electrical current corresponds to about
55-60 °C, i.e. a temperature at which DA processes between decoupled furan/maleimide
moieties occur as demonstrated for thermal healing using a conventional oven [45, 47, 60].
The sequence of cross-linking in solution, de-cross-linking during molding and healing by
resistive heating of the nanocomposite was also monitored by ATR-FTIR spectroscopy
(Figure 6.11). The normalized spectra at 2930 cm-1
(C-H stretch, Figure 6.11A) and at 1145
cm-1
(C-N stretching band [61], Figure 6.11B) evidence the intensity variations of the C-N-C
peak at 1185 cm-1
attributed to the stretching of the succinimide ring in the DA adduct). It is
worth noting that the DA band decreased in intensity after molding (decoupling of DA
adducts), whereas it promptly recovered after healing to intensities similarly collected form
the nanocomposite thermally cross-linked in solution.
A B C
111
Figure 6.11 ATR-FTIR spectra of PK-Fu cross-linked with b-Ma and reinforced with MWCNTs. The
different colors refer to the processes of: cross-linking (DA) (black), molding (partial rupture of DA
adducts or r-DA) (red) and healing by resistive heating (reconnection of DA adducts) (blue).
According to the results displayed in Figure 6.11, part of the DA adducts are broken after
molding, which could possibly favour re-agglomeration of MWCNTs. To support this
hypothesis, resistive measurements and morphological studies by SEM were combined in
order to figure out the degree of dispersion of MWCNTs in the matrix (Table 6.3, Figure
6.12).
Higher surface resistivity (about 228 Ω/cm2) was observed for the compressed bar after
molding as compared with the one healed by 24 h of resistive heating (about 206 Ω/cm2).
This result suggests the potentiality offered by the resistive heating in providing a better
percolative network (Table 6.3) [62].
Table 6.3 Electrical parameters and surface resistivity (ρs) results for the thermoset composite after
molding and healing by resistive heating at 35 V.
Parameters After molding
After healing
resistive heating Current (A) 30.7 ± 1.2 33.9 ± 0.3
ρs (Ω/cm2) 228.02 ± 9.5 206.3 ± 1.8
1800 1600 1400 1200 1000 800
6
5
4
3
2
1
0
1220 1200 1180 1160 1140 1120 11001,25
1,20
1,15
1,10
1,05
1,00
0,95
0,90
Diels-Alder at 50 oC in THF
Molding at 150 oC (r-DA)
After healing
Tra
nsm
itta
nce
Wavenumber (cm-1)
Ab
sorb
ance
(-)
Wavenumber (cm-1)
112
Figure 6.12 SEM micrographs of PK-Fu/b-Ma/MWCNTs composite after molding (A, C) and after
healing by resistive heating (B, D).
Contrast mode SEM micrographs [63] corroborate the results obtained from electrical
measurements on the same sample after being molded and successively healed by resistive
heating (Figure 6.12). On the freshly broken surfaces of the composite after molding (Figure
6.12A, 6.12C), bundles of MWCNTs (indicated by white arrows) can be observed.
Conversely, a better distribution of MWCNTs is present after resistive heating (Figures
6.12B, 6.12D), thus confirming the ability of the electric current in restoring a more effective
percolative network.
6.4 Conclusions
We designed an intrinsic self-healing thermoset/MWCNT nanocomposite capable to recover
structural damage by means of thermally reversible Diels-Alder links activated by resistive
heating. The thermoset matrix demonstrated to chemically interact with the filler due to the
diene/dienophile character of MWCNTs. By this approach, it was possible to realize a
113
thermoset nanocomposite displaying a relatively high softening point (Tan δ) and electrical
conductive properties. Thermo-mechanical measurements as well as spectroscopy and
microscopy analysis clearly established the reworkability and intrinsic self-healing character
of the thermoset nanocomposite via Diels-Alder chemistry using resistive heating as external
stimulus. In addition, the industrial scale by which PK can be produced and chemically
modified offers a straightforward alternative for the industrial production of electrically
conductive thermoset nanocomposites displaying self-healing ability by resistive heating
activation of reversible Diels-Alder adducts.
6.5 References
1. Ratna D. Handbook of thermoset resins. United Kingdom: iSmithers, 2009.
2. Delehanty JM, Moss I, Westall S, Dowden T. Room temperature curable silicone
elastomer composition;EP20070824749(EP2089465 B1).
3. Zhang MQ, Rong MZ, editors. Self-Healing Polymers and Polymer Composites. Hoboken,
New Jersey.: Jhon Wiley and Sons, Inc., 2011.
4. Zhang MQ, Rong MZ. Intrinsic self-healing of covalent polymers through bond
reconnection towards strength restoration. Polymer Chemistry 2013;4(18):4878-4884.
5. Ward IM, Sweeney J. An introduction to the mechanical properties of solid polymers.
Chichester, England: John Wiley and Sons Ltd, 2004.
6. Nakamura Y, Yamaguchi M, Okubo M, Matsumoto T. Effect of particle-size on the
fracture-toughness of epoxy-resin filled with spherical silica. Polymer 1992;33(16):3415-
3426.
7. Lee J, Yee AF. Inorganic particle toughening II: toughening mechanisms of glass bead
filled epoxies. Polymer 2001;42(2):589-597.
8. Pearson RA, Yee AF. Toughening mechanisms in thermoplastic-modified epoxies. 1.
Modification using poly(phenylene oxide. Polymer 1993;34(17):3658-3670.
9. Rusli A, Cook WD, Schiller TL. Blends of epoxy resins and polyphenylene oxide as
processing aids and toughening agents 2: Curing kinetics, rheology, structure and properties.
Polym Int 2014;63(8):1414-1426.
10. Giannakopoulos G, Masania K, Taylor AC. Toughening of epoxy using core-shell
particles. J Mater Sci 2011;46(2):327-338.
11. Chen J, Kinloch AJ, Sprenger S, Taylor AC. The mechanical properties and toughening
mechanisms of an epoxy polymer modified with polysiloxane-based core-shell particles.
Polymer 2013;54(16):4276-4289.
12. Kinloch AJ, Lee SH, Taylor AC. Improving the fracture toughness and the cyclic-fatigue
resistance of epoxy-polymer blends. Polymer 2014;55(24):6325-6334.
13. Yadav SN, Kumar V, Verma SK. Fracture toughness behaviour of carbon fibre epoxy
composite with Kevlar reinforced interleave. Materials Science and Engineering B-Solid
State Materials for Advanced Technology 2006;132(1-2):108-112.
14. Pimenta S, Pinho ST. Recycling carbon fibre reinforced polymers for structural
applications: Technology review and market outlook. Waste Manage 2011;31(2):378-392.
15. Wu G, Liu D, Liu G, Chen J, Huo S, Kong Z. Thermoset nanocomposites from
waterborne bio-based epoxy resin and cellulose nanowhiskers. Carbohydr Polym
2015;127:229-235.
114
16. Feldman D. Polyblend Nanocomposites. Journal of Macromolecular Science Part A-Pure
and Applied Chemistry 2015;52(8):648-658.
17. McNally, Tony & Pötschke, Petra. Polymer-carbon nanotube composites. Philadelphia,
USA: Woodhead Publishing Limited, 2011.
18. Leng J, Kin-Tak Lau A. Multifunctional Polymer Nanocomposites. Boca Raton, London,
New York: Taylor and Francis Group, 2011.
19. White S, Sottos N, Geubelle P, Moore J, Kessler M, Sriram S, Brown E, Viswanathan S.
Autonomic healing of polymer composites. Nature 2001;409(6822):794-797.
20. McNally, Tony & Pötschke, Petra. Preparation and processing of polymer-carbon
nanotube composites. In: Cambridge, UK: Woodhead Publishing Limited, 2011. p. 3.
21. Calisi N, Giuliani A, Alderighi M, Schnorr JM, Swager TM, Di Francesco F, Pucci A.
Factors affecting the dispersion of MWCNTs in electrically conducting SEBS
nanocomposites. European Polymer Journal 2013;49(6):1471-1478.
22. Koval'chuk AA, Shevchenko VG, Shchegolikhin AN, Nedorezova PM, Klyamkina AN,
Aladyshev AM. Effect of Carbon Nanotube Functionalization on the Structural and
Mechanical Properties of Polypropylene/MWCNT Composites. Macromolecules
2008;41(20):7536-7542.
23. Williams GA, Ishige R, Cromwell OR, Chung J, Takahara A, Guan Z. Mechanically
Robust and Self-Healable Superlattice Nanocomposites by Self-Assembly of Single-
Component "Sticky" Polymer-Grafted Nanoparticles. Advanced materials (Deerfield Beach,
Fla.) 2015;27(26):3934-3941.
24. Ahmed AS, Ramanujan RV. Curie temperature controlled self-healing magnet-polymer
composites. J Mater Res 2015;30(7):946-958.
25. Crespy D, Zhao Y. Preparation of nanocapsules and core-shell nanofibers for extrinsic
self-healing materials. In: Binder WH, editor. Weinheim, Germany: Wiley-VCH Verlag
GmbH and Co., 2013. p. 822.
26. Yuan YC, Yin T, Rong MZ, Zhang MQ. Self healing in polymers and polymer
composites. Concepts, realization and outlook: A review. Express Polym Lett 2008;2(4):238-
250.
27. Li G. Self-healing composites: Shape Memory Polymer Based Structures. United
Kingdom: Jhon Wiley & Sons Ltd, 2015.
28. Alessandri I. Self-Healing Nanocomposites: role and activation of inorganic moieties and
hybrid nanophases. In: Amendola V, Meneghetti M, editors. Boca raton, London, New York:
CRC press, Taylor and Francis Group, 2012. p. 163.
29. Zhong N, Post W. Self-repair of structural and functional composites with intrinsically
self-healing polymer matrices: A review. Composites Part A-Applied Science and
Manufacturing 2015;69:226-239.
30. Long KN. The mechanics of network polymers with thermally reversible linkages. J
Mech Phys Solids 2014;63:386-411.
31. Burnworth M, Tang L, Kumpfer JR, Duncan AJ, Beyer FL, Fiore GL, Rowan SJ, Weder
C. Optically healable supramolecular polymers. Nature 2011;472(7343):334-338.
32. Toncelli C, De Reus D, Broekhuis AA, Picchioni F. Thermoreversibility in Polymeric
Systems. In: Amendola M, Meneghetti V, editors. New York: CRC Press, 2011. p. 199.
33. Khor SP, Varley RJ, Shen SZ, Yuan Q. Thermo-reversible healing in a crosslinked
polymer network containing covalent and thermo-reversible bonds. J Appl Polym Sci
2013;128(6):3743-3750.
34. Skene WG, Lehn JMP. Dynamers: Polyacylhydrazone reversible covalent polymers,
component exchange, and constitutional diversity. Proc Natl Acad Sci U S A
2004;101(22):8270-8275.
115
35. Kuang X, Liu G, Dong X, Wang D. Enhancement of Mechanical and Self-Healing
Performance in Multiwall Carbon Nanotube/Rubber Composites via Diels-Alder Bonding.
Macromolecular Materials and Engineering 2016;301(5):535-541.
36. Yang Y, Zhu B, Yin D, Wei J, Wang Z, Xiong R, Shi J, Liu Z, Lei Q. Flexible self-
healing nanocomposites for recoverable motion sensor. Nano Energy 2015;17:1-9.
37. Hohlbein N, Shaaban A, Schmidt AM. Remote-controlled activation of self-healing
behavior in magneto-responsive ionomeric composites. Polymer 2015;69:301-309.
38. Li H, Li H, Li Z, Lin F, Liu D, Wang W, Wang B, Xu Z. T pattern fuse construction in
segment metallized film capacitors based on self-healing characteristics. Microelectronics
Reliability 2015;55(6):945-951.
39. Chuo T, Wei T, Liu Y. Electrically driven self-healing polymers based on reversible
guest-host complexation of -cyclodextrin and ferrocene. Journal of Polymer Science Part A-
Polymer Chemistry 2013;51(16):3395-3403.
40. Swait TJ, Rauf A, Grainger R, Bailey PBS, Lafferty AD, Fleet EJ, Hand RJ, Hayes SA.
Smart composite materials for self-sensing and self-healing. Plastics Rubber and Composites
2012;41(4-5):215-224.
41. Wang CC, Ding Z, Purnawali H, Huang WM, Fan H, Sun L. Repeated Instant Self-
healing Shape Memory Composites. Journal of Materials Engineering and Performance
2012;21(12):2663-2669.
42. Cui HP, Song CL, Huang WM, Wang CC, Zhao Y. Rubber-like electrically conductive
polymeric materials with shape memory. Smart Mater Struct 2013;22(5):055024.
43. Willocq B, Bose RK, Khelifa F, Garcia SJ, Dubois P, Raquez J-. Healing by the Joule
effect of electrically conductive poly(ester-urethane)/carbon nanotube nanocomposites.
Journal of Materials Chemistry a 2016;4(11):4089-4097.
44. Kim J, Sauti G, Siochi EJ, Smith JG, Wincheski RA, Cano RJ, Connell JW, Wise KE.
Toward High Performance Thermoset/Carbon Nanotube Sheet Nanocomposites via Resistive
Heating Assisted Infiltration and Cure. Acs Applied Materials & Interfaces
2014;6(21):18832-18843.
45. Lee J, Stein IY, Kessler SS, Wardle BL. Aligned Carbon Nanotube Film Enables
Thermally Induced State Transformations in Layered Polymeric Materials. Acs Applied
Materials & Interfaces 2015;7(16):8900-8905.
46. Araya-Hermosilla R, Broekhuis AA, Picchioni F. Reversible polymer networks
containing covalent and hydrogen bonding interactions. European Polymer Journal
2014;50:127-134.
47. Zhang Y, Broekhuis AA, Picchioni F. Thermally Self-Healing Polymeric Materials: The
Next Step to Recycling Thermoset Polymers?. Macromolecules 2009;42:1906-1912.
48. Toncelli C, De Reus DC, Picchioni F, Broekhuis AA. Properties of Reversible Diels-
Alder Furan/Maleimide Polymer Networks as Function of Crosslink Density.
Macromolecular Chemistry and Physics 2012;213(2):157-165.
49. Mul WP, Dirkzwager H, Broekhuis AA, Heeres HJ, van der Linden AJ, Orpen AG.
Highly active, recyclable catalyst for the manufacture of viscous, low molecular weight, CO-
ethene-propene-based polyketone, base component for a new class of resins. Inorg Chim Acta
2002;327:147-159.
50. Zhang Y, Broekhuis AA, Stuart MCA, Picchioni F. Polymeric Amines by chemical
modifications of alternating aliphatic polyketones. J Appl Polym Sci 2008;107(1):262-271.
51. LISE laboratory, Namur University, Belgium. Winspec software.
52. Zydziak N, Huebner C, Bruns M, Barner-Kowollik C. One-Step Functionalization of
Single-Walled Carbon Nanotubes (SWCNTs) with Cyclopentadienyl-Capped
Macromolecules via Diels-Alder Chemistry. Macromolecules 2011;44(9):3374-3380.
116
53. Prado LASdA, Kwiatkowska M, Funari SS, Roslaniec Z, Broza G, Schulte K. Studies on
Morphology and Interphase of Poly(butylene terephthalate)/Carbon Nanotubes
Nanocomposites. Polym Eng Sci 2010;50(8):1571-1576.
54. Meuer S, Braun L, Schilling T, Zentel R. alpha-Pyrene polymer functionalized
multiwalled carbon nanotubes: Solubility, stability and depletion phenomena. Polymer
2009;50(1):154-160.
55. Lee J, Yoon S, Kim KK, Cha I, Park YJ, Choi J, Lee YH, Paik U. Exfoliation of single-
walled carbon nanotubes induced by the structural effect of perylene derivatives and their
optoelectronic properties. Journal of Physical Chemistry C 2008;112(39):15267-15273.
56. Chang C, Liu Y. Functionalization of multi-walled carbon nanotubes with furan and
maleimide compounds through Diels-Alder cycloaddition. Carbon 2009;47(13):3041-3049.
57. R. Araya-Hermosilla, G. Fortunato, A. Pucci, A. A. Broekhuis, P. Raffa, G. M. R Lima,
P. Pourhossein, L. Polgar and F. Picchioni. Thermally reversible rubber-toughened thermoset
networks via Diels-Alder chemistry. Submitted to European Polymer Journal.
58. Mauro M, Acocella MR, Corcione CE, Maffezzoli A, Guerra G. Catalytic activity of
graphite-based nanofillers on cure reaction of epoxy resins. Polymer 2014;55(22):5612-5615.
59. Park JS, Darlington T, Starr AF, Takahashi K, Riendeau J, Thomas Hahn H. Multiple
healing effect of thermally activated self-healing composites based on Diels–Alder reaction.
Composites Sci Technol 2010;70(15):2154-2159.
60. Edward D. Sosa, Thomas K. Darlington, Brian A. Hanos, and Mary Jane E. O’Rourke.
Multifunctional Thermally Remendable Nanocomposites. Journal of Composites 2014;2014.
61. Silverstein RM, Bassler GC, Morrill TC. Spectrometric identification of organic
compounds. Canada: John Wiley and sons, Inc., 1991.
62. Alig I, Skipa T, Lellinger D, Bierdel M, Meyer H. Dynamic percolation of carbon
nanotube agglomerates in a polymer matrix: comparison of different model approaches.
Physica Status Solidi B-Basic Solid State Physics 2008;245(10):2264-2267.
63. Deng H, Skipa T, Bilotti E, Zhang R, Lellinger D, Mezzo L, Fu Q, Alig I, Peijs T.
Preparation of High-Performance Conductive Polymer Fibers through Morphological Control
of Networks Formed by Nanofillers. Advanced Functional Materials 2010;20(9):1424-1432.
117
7 Appendix
7.1 Summary
Over the last few decades, researchers working on the design and optimization of plastic
materials have reported considerable improvements (e.g. in mechanical properties of recycled
materials) often stimulated by societal driven demands in terms of sustainability,
enhancement of material quality and value-add services. However, despite all the efforts, it is
widely recognized that plastic recycling is still facing relevant problems regarding collection,
separation, cleaning, processing chemistry, and flow markets for recycled products. In the
particular case of thermoplastics, suitable materials for recycling are subjected to the
availability of treatment technology (e.g. re-melting, re-shaping) and markets. Unlike
thermoplastics, thermosetting resins are polymers that upon heat or light radiation convert
into infusible and insoluble polymer networks being cross-linked by covalent chemical
bonds. Due to this network, thermosets offer many superior properties compared to
thermoplastics such as mechanical strength, dimensional stability at elevated temperature and
solvent resistance. These features confer them the ability to be widely used as composites,
adhesives, and coatings. However, thermosets cannot be re-melted or re-shaped, since they
degrade or decompose upon heating or by chemical treatments. These features limit their
recyclability so that at the end of their service life, they are simply grinded to produce cheap
reinforcement fillers or are thermally processed (pyrolysis, incineration) to recover the fiber
content or just energy. A more sustainable strategy is to re-design thermosets by replacing
classical chemical cross-links by reversible covalent bonds that can be repeatedly broken and
re-connected by heating. These thermoreversible networks combine many mechanical
properties of thermosets with the processability of thermoplastics. In addition, they can be
designed to provide intrinsic self-healing properties.
In this thesis, chapters 2 and 3 focused on the chemical modification of alternating aliphatic
polyketone with aliphatic and aromatic amine compounds using the Paal-Knorr reaction to
obtain thermally reversible polymers with relatively high glass transition temperatures (i.e. a
Tg window from 100 °C to 185 °C). Both chapters highlight the tuneable thermal properties
of polymers containing furan (Diels-Alder) and hydrogen bonding active groups directly
attached to the backbone. The reversible thermosetting of furan-functionalized polymers via a
Diels-Alder and retro-Diels-Alder sequence with bis-maleimide was investigated and
118
systematically compared with thermoset polymers containing both reversible covalent and
supramolecular interactions.
Chapter 4 reports on the preparation of a reversible and toughened thermoset system based on
the covalent incorporation of furan-functionalized ethylene-propylene rubber into a thermoset
furan-functionalized polyketone. Spectroscopy and thermal analyses proved the reversible
interaction between the functionalized polymers with b-Ma via DA and r-DA sequence.
Likewise, thermo-mechanical experiments indicated the reworkability of the material with no
evident differences in elastic and loss modulus after several heating cycles and recycling
steps. Moreover, a considerable increase in the softening point (also known as damping
factor) was also found for the toughened system containing the furan-functionalized rubber.
A two-fold increase in Izod impact strength compared to the original thermoset was also
recorded for the toughened system.
Chapter 5 focused on the design of an engineered thermoplastic polymer containing pyrrole
units in the main chain and hydroxyl pendant groups, which help in achieving
nanocomposites containing well-distributed, exfoliated and undamaged MWCNTs. The
incorporation of MWCNTs made it possible to change the material from an insulator to a
conductive system displaying temperature sensor properties. Notably, the resistivity–
temperature profile is very reproducible which suggests the potential application of the
composite as a temperature sensor.
Chapter 6 describes the electrically-induced self-healing properties of a thermoset
nanocomposite designed by mixing furan-functionalized polyketone cross-linked with
aromatic bis-maleimide and MWCNTs via Diels-Alder (DA) reversible cycloaddition. The
incorporation of MWCNTs increases the material modulus and allowed electrical and thermal
conduction. XPS analysis indicates that the MWCNTs, due to their diene/dienophile
character, covalently interact with both furan and maleimide groups via Diels-Alder reaction,
leading to good interfacial adhesion between filler and matrix. It is also demonstrated that
above the thermal stability of the composite, the retro-DA process is triggered disrupting the
network. Hence, the mechanical properties of the material are diminished due to the
decoupling of DA adducts. Nevertheless, the mechanical properties are completely recovered
after the sample is healed through electrical resistive heating.
119
7.2 Samenvatting
In de afgelopen decennia hebben onderzoekers die aan het ontwerp en de optimalisatie van
kunststoffen werken aanzienlijke verbeteringen gerapporteerd (bijv. in de mechanische
eigenschappen van gerecyclede materialen). Deze verbeteringen zijn vaak gestimuleerd door
eisen vanuit de maatschappij op het gebied van duurzaamheid en verbetering van de kwaliteit
van het materiaal. Echter, ondanks alle inspanningen is het is algemeen erkend dat plastic
recycling nog steeds onderhevig is aan relevante problemen met betrekking tot inzameling,
scheiding, schoonmaken, verwerken van de chemie en de markten voor de gerecyclede
producten. In het specifieke geval van thermoplasten zijn geschikte materialen voor recycling
onderhevig aan de beschikbaarheid van behandelingstechnologie (bijvoorbeeld hersmelten en
hervormen) en markten. Anders dan bij thermoplasten worden thermohardende harsen na een
behandeling met warmte of licht omgezet in onsmeltbare en onoplosbare polymere netwerken
die met covalente chemische bindingen vernet zijn. Dit netwerk geeft thermoharders duidelijk
betere eigenschappen vergeleken met thermoplasten. Dit heeft met name betrekking op de
mechanische kracht, vormvastheid bij hoge temperatuur en bestendigheid tegen
oplosmiddelen. Deze eigenschappen zorgen er voor dat ze op grote schaal worden gebruikt
als composieten, hechtmiddelen en coatings. Thermoharders kunnen echter niet opnieuw
worden gesmolten en opnieuw worden vormgegeven omdat ze degraderen of ontleden bij
verwarming of door chemische behandelingen. Dit beperkt hun recycleerbaarheid en zorgt er
voor dat ze einde van hun levensduur gewoon vermalen worden tot goedkope vulstoffen of
thermisch behandeld worden (dmv pyrolyse of verbranding) om de vezels of ten minste wat
energie terug te winnen. Een meer duurzame strategie is gevonden in het herontwerpen van
thermoharders door de klassieke netwerkchemie te vervangen met door omkeerbare covalente
bindingen die herhaaldelijk kunnen worden afgebroken en opnieuw met elkaar verbonden
kunnben worden door ze te verwarmen. Deze thermo-reversibele netwerken combineren veel
mechanische eigenschappen van thermoharders met de verwerkbaarheid van thermoplasten.
Bovendien kunnen ze worden ontworpen om voor intrinsieke zelfherstellende eigenschappen
van materialen te zorgen.
Hoofdstukken 2 en 3 van dit proefschrift zij gericht op de chemische modificatie van
alternerende alifatische polyketonen met alifatische en aromatische amineverbindingen via de
Paal-Knorr reactie om thermisch reversibele polymeren met relatief hoge glasovergangs-
temperaturen (een Tg venster van 100 °C tot 185 °C) te verkrijgen. Beide hoofdstukken
120
markeren de afstembaarheid van de thermische eigenschappen van polymeren met furan
(Diels-Alder) en waterstofbindende actieve groepen die direct aan de hoofdketen bevestigd
zijn. Het omkeerbare vernetten van furan-gefunctionaliseerde polymeren via de Diels-Alder
en retro-Diels-Alder reacties met bis-maleïmide zijn systematisch onderzocht en vergeleken
met thermohardende polymeren die zowel omkeerbare covalente als supramoleculaire
interacties hebben.
Hoofdstuk 4 beschrijft de bereiding van een reversibele en geharde vesterkte thermoharder
die gebaseerd is op de covalente interactie tussen furan-gefunctionaliseerde etheen-propeen
rubbers en een matrix van furan-gefunctionaliseerd polyketon. Spectroscopie en thermische
analyse werden gebruikt om de reversibele interactie tussen de gefunctionaliseerde
polymeren met bismaleimide via DA en de retro-DA reactie te laten zien. Eveneens werden
thermomechanische experimenten gebruikt om de her-verwerkbaarheid van het materiaal
zonder duidelijke verschillen in elastische en verliesmodulus na verschillende
verwarmingsperioden en recycling stappen te demonstreren. Bovendien werd een
aanzienlijke toename van de verwekingstemperatuur (ook bekend als dempingsfactor)
gevonden voor het systeem dat versterkt was met het furan-gefunctionaliseerde rubber. Een
verdubbeling van de Izod slagsterkte ten opzichte van de oorspronkelijke thermoharder werd
ook verkregen voor het met rubber versterkte systeem.
Hoofdstuk 5 is gericht op het ontwerp van een thermoplastische polymeer met pyrrool
eenheden in de hoofdketen en hydroxyl zijgroepen die helpen bij het mengen van ontvouwen
en onbeschadigde MWCNTs in nanocomposieten. De inmengen van MWCNTs maakte het
mogelijk om het materiaal van een isolator in een geleidende systeem met temperatuursensor
eigenschappen te veranderen. Met name het weerstand-temperatuur profiel is van deze
materialen is zeer reproduceerbaar en nodigt uit om deze composiet in een toepassing als
temperatuursensor te gebruiken.
Hoofdstuk 6 beschrijft de elektrisch geïnduceerde zelfherstellende eigenschappen van een
thermohardent nanocomposiet dat gemaakt is door furan-gefunctionaliseerde polyketonen te
vernetten met aromatische bis-maleïmides en MWCNTs via de reversibele Diels-Alder (DA)
cycloadditie. De opname van MWCNTs verhoogt de modulus van het materiaal en zorgt voor
elektrische en thermische geleiding. XPS-analyse laat zien dat de MWCNTs, vanwege hun
dieen/diënofiel karakter, een covalente interactie met zowel furan als maleïmidegroepen
hebben via Diels-Alder-reactie. Hierdoor ontstaat er een goede grensvlak hechting tussen de
121
vulstof en de polymere matrix. Het is ook aangetoond dat een gegeven temperatuur, de retro-
DA proces het netwerk verstoort. Zodoende worden de mechanische eigenschappen van het
materiaal verminderd door het ontkoppelen van de DA adducten. Desalniettemin zijn de
mechanische eigenschappen volledig hersteld nadat het monster is gerepareerd door middel
van elektrische weerstandsverwarming.
122
7.3 List of publications
1. R. Araya-Hermosilla, A. Pucci, E. Araya-Hermosilla, P. P. Pescarmona, P. Raffa, L.
M. Polgar, I. Moreno-Villoslada, M. Flores, G. Fortunato, A. A. Broekhuis and F.
Picchioni. An easy synthetic way to exfoliate and stabilize MWCNTs in a
thermoplastic pyrrole-containing matrix assisted by hydrogen bonds (2016). RSC
Advances, 6, 85829-85837.
2. L. M. Polgar, G. Fortunato, R. Araya-Hermosilla, M. van Duin, A. Pucci, F.
Picchioni. Cross-linking of rubber in the presence of multi-functional cross-linking
aids via thermoreversible Diels-Alder chemistry (2016). European Polymer Journal,
82, 208–219.
3. R. Araya-Hermosilla, G. M. R Lima, I. Moreno-Villoslada, M. Flores, A. A.
Broekhuis and F. Picchioni. Intrinsic self-healing thermoset through covalent and
hydrogen bonding interactions (2016). European Polymer Journal, 81, 186–197.
4. R. Araya-Hermosilla, G. Fortunato, A. Pucci, P. Raffa, L. Polgar, A. A. Broekhuis, P.
Pourhossein, G. M. R Lima, M. Beljaars and F. Picchioni. Thermally reversible
rubber-toughened thermoset networks via Diels-Alder chemistry (2015). European
Polymer Journal, 74, 229–240.
5. S. Orellana, C. Torres-Gallegos R. Araya-Hermosilla, F. Oyarzun-Ampuero, I.
Moreno-Villoslada. Association Efficiency of Three Ionic Forms of Oxytetracycline
to Cationic and Anionic Oil In-Water Nanoemulsions Analyzed by Diafiltration
(2015). Journal of Pharmaceutical Sciences, 104, 3, 1141–1152.
6. M. E. Flores, Naoki Sano, R. Araya-Hermosilla, Toshimichi Shibue, A. F. Olea,
Hiroyuki Nishide, I. Moreno-Villoslada. Self-association of 5,10,15,20-tetrakis-(4-
sulfonatophenyl)-porphyrin tuned by poly(decylviologen) and sulfobutylether-β-
cyclodextrin (2015). Dyes and Pigments, 112, 262-273.
7. R. Araya-Hermosilla, A.A. Broekhuis, F. Picchioni. Reversible polymer networks
containing covalent and hydrogen bonding interactions (2014). European Polymer
Journal, 50, 127-134.
8. R. Araya-Hermosilla, E. Araya-Hermosilla, C. Torres-Gallegos, C. Alarcón-Alarcón,
Hiroyuki Nishide, I. Moreno-Villoslada. Sensing Cu2+ by controlling the aggregation
properties of the fluorescent dye rhodamine 6G with the aid of polyelectrolytes
bearing different linear aromatic density (2013). Reactive and Functional Polymers,
73, 11, 1455-1463.
123
7.4 Contribution to international conferences
1. Lorenzo Massimo Polgar, Rodrigo Araya-Hermosilla, Machiel van Essen, Mattia
Lenti, Martin van Duin, Francesco Picchioni (4-6 March 2016). Thermo-reversibly
cross-linked rubbers and self-healing rubber compounds. WCMS2016, Singapore.
Oral presentation.
2. Lorenzo M. Polgar, G. Fortunato, M. Lenti, M. van Essen, R. Araya-Hermosilla, M.
van Duin, Francesco Picchioni (29 November – 2 December 2015). Conductive,
self-healing cross-linked rubber compounds. CHAINS, Veldhoven, the
Netherlands. Oral presentation.
3. Lorenzo M. Polgar, G. Fortunato, M. Lenti, R. Araya-Hermosilla, M. van Duin,
Francesco Picchioni (3-4 November 2015). Reworkable rubbers and self-healing
rubber compounds. Dutch Polymer Institute Annual Meeting, the Netherlands. Oral
presentation.
4. A. Pucci, R. Araya-Hermosilla, P. Raffa, G. Fortunato, A. A. Broekhuis and F.
Picchioni (September 2015). Self-healing and electronic conducting properties of
a CNT/polymer composite. Congresso Divisione Chimica Industriale. Salerno, Italy.
Oral presentation.
5. L.M. Polgar, M. Lenti, M. van Duin, R. Araya-Hermosilla, P. Raffa, A.A. Broekhuis
and F. Picchioni (22-24 June 2015). Intrinsic self-healing of thermoreversibly
cross-linked rubber compounds. ICSHM2015, Durham, United States of America.
Oral presentation.
6. R. Araya-Hermosilla, P. Raffa, A. Pucci, G. Fortunato, A. A. Broekhuis and F.
Picchioni (9-13 march 2015). Self-healing and conductive properties of a
CNT/polymer composite. Fourth International Conference on Multifunctional,
Hybrid and Nanomaterials. Barcelona, Spain. Oral presentation.
7. R. Araya-Hermosilla, C. Toncelli and F. Picchioni (2012). Metal ions uptake by
polyketone chemically modified by Paal-Knorr reaction. ANQUE International
Congress of Chemical Engineering. Sevilla, Spain. Poster.
124
7.5 Acknowledgements
My biggest gratitude is to my parents, Mercedes and Valdemar, who have sacrificed their
own life for my education and the education of my brother. My biggest thankfulness is to my
wife for supporting me during more than eleven years and giving me the most beautiful thing
that someone can give to you, another life! (my daughter Gabriela). I thank my brother
Esteban for his support during all these years that we have been studying so far from home.
I would like to express my sincere gratitude to my supervisor Prof. dr. Francesco Picchioni
for his continuous support during my PhD, for his motivation during bad times, and the
immense knowledge transfer. His guidance helped me during all this time of research and
writing of this thesis. Since the beginning of my PhD he showed himself not only as a
supervisor, but also as a friend. Most of all, I thank him his reliance over my capacities. It
was a risky bet: to trust in a marine biologist to finish a PhD in a chemical engineering
environment.
I would like to thank the reading committee of my thesis: Prof. dr. Katja Loos, Prof. dr. H. J.
(Erik) Heeres, and Prof. dr. Eduardo Soto for their insightful comments and encouragement
to widen my research from various perspectives.
My sincere thanks also goes to Prof. dr. Andrea Pucci for his friendship and enormous
transfer of knowledge during my research work and writing of scientific articles. Likewise, I
would like to thank professors: dr. Paolo Pescarmona, dr. Petra Rudolf, dr. A. A. (Ton)
Broekhuis, dr. Ignacio Moreno Villoslada and dr. Patrizio Raffa for their insightful comments
and enlightening advices during my research work and writing of scientific articles.
My sincere thanks also goes to Marcel de Vries, Anne Appledoorn and Erwin Wilbers for all
their help to resolve technical issues during my work in the laboratory. Likewise, I want to
thank to Jan Henk Marsman and Léon Rohrbach for their help with the analytical
measurements during all my PhD research. Also, I want to thank to Gert Alberda van
Ekenstein for his insightful advices and help during thermal and mechanical characterization.
I would like to thank to Professor dr. Václav Ocelik for his insightful advices regarding
microscopy characterization. My special thanks goes to Marya de Jonge who solved all my
bureaucratic problems. Most of all, I appreciate her patience to teach me het ongelofelijk
moeilijk taal van Netherlands. I also thank Annette Korringa for her help regarding the
arrangement of bureaucratic issues of my scholarship and university administrations.
125
All the work here presented could not be possible to conduct without the help of all my
collaborators and colleagues. My sincere thanks goes to Guilherme Rooweder Lima,
Giovanni Fortunato, Dian Santosa, dr. Mario Flores, dr. Parisa Pourhossein, Martijn Beljaars,
Machiel van Essen, Mattia Lenti and Régis Gengler. My special thanks goes to my friends
and colleagues dr. Patrizio Raffa, dr. Diego Wever, Lorenzo Polgar (many thanks for your
help with all Dutch translations of my thesis), Graham Ramalho, Pablo Druetta, Patrick
Figaroa, Arjen Kamphuis, Bilal Niazi, Eric Benjamins and Arne Hommes with who I could
mix very funny moments and work at the same time.
My sincere thanks goes also to my dearest friends Jonathan Pothuis, Sarah Ashworth,
Cristian Santana, dr. Andréa Medina, dr. Claudio Toncelli and Benny Backer who in one way
or another changed the course of my life.
I am also grateful to the support of the Programa Formación de Capital Humano Avanzado,
CONICYT, BECAS CHILE; grant number: 72111428.
Finally, I want to thank to all people who work in the Product Technology Department and
Green Chemical Reaction Engineering Department belonging to the Engineering and
Technology institute of Groningen (ENTEG).
Rodrigo Andrés Araya Hermosilla